From 8e5301e4ed16248f9f46a3dae2df0dc037cb24ed Mon Sep 17 00:00:00 2001 From: jwu049 Date: Sun, 11 Aug 2024 23:00:36 +0800 Subject: [PATCH] add the code of ch07 --- Codes/ch07/01_main-chapter-code/README.md | 76 + Codes/ch07/01_main-chapter-code/ch07.ipynb | 2795 ++++++ .../exercise-solutions.ipynb | 1002 +++ .../exercise_experiments.py | 562 ++ .../ch07/01_main-chapter-code/gpt_download.py | 142 + .../gpt_instruction_finetuning.py | 351 + .../instruction-data-with-response.json | 662 ++ .../instruction-data.json | 5502 ++++++++++++ .../load-finetuned-model.ipynb | 219 + .../01_main-chapter-code/ollama_evaluate.py | 126 + .../01_main-chapter-code/previous_chapters.py | 470 + Codes/ch07/01_main-chapter-code/tests.py | 16 + Codes/ch07/02_dataset-utilities/README.md | 80 + Codes/ch07/02_dataset-utilities/config.json | 4 + .../create-passive-voice-entries.ipynb | 426 + .../find-near-duplicates.py | 153 + .../instruction-examples-modified.json | 1202 +++ .../instruction-examples.json | 1002 +++ .../requirements-extra.txt | 3 + Codes/ch07/03_model-evaluation/README.md | 26 + Codes/ch07/03_model-evaluation/config.json | 4 + .../eval-example-data.json | 702 ++ .../llm-instruction-eval-ollama.ipynb | 648 ++ .../llm-instruction-eval-openai.ipynb | 571 ++ .../requirements-extra.txt | 2 + .../scores/correlation-analysis.ipynb | 269 + .../scores/gpt4-model-1-response.json | 1 + .../scores/gpt4-model-2-response.json | 1 + .../scores/llama3-8b-model-1-response.json | 1 + .../scores/llama3-8b-model-2-response.json | 1 + .../04_preference-tuning-with-dpo/README.md | 7 + .../create-preference-data-ollama.ipynb | 595 ++ .../dpo-from-scratch.ipynb | 3096 +++++++ .../instruction-data-with-preference.json | 7702 +++++++++++++++++ .../previous_chapters.py | 470 + Codes/ch07/05_dataset-generation/README.md | 6 + .../instruction-data-llama3-7b.json | 22 + .../05_dataset-generation/llama3-ollama.ipynb | 506 ++ Codes/ch07/README.md | 15 + 39 files changed, 29438 insertions(+) create mode 100644 Codes/ch07/01_main-chapter-code/README.md create mode 100644 Codes/ch07/01_main-chapter-code/ch07.ipynb create mode 100644 Codes/ch07/01_main-chapter-code/exercise-solutions.ipynb create mode 100644 Codes/ch07/01_main-chapter-code/exercise_experiments.py create mode 100644 Codes/ch07/01_main-chapter-code/gpt_download.py create mode 100644 Codes/ch07/01_main-chapter-code/gpt_instruction_finetuning.py create mode 100644 Codes/ch07/01_main-chapter-code/instruction-data-with-response.json create mode 100644 Codes/ch07/01_main-chapter-code/instruction-data.json create mode 100644 Codes/ch07/01_main-chapter-code/load-finetuned-model.ipynb create mode 100644 Codes/ch07/01_main-chapter-code/ollama_evaluate.py create mode 100644 Codes/ch07/01_main-chapter-code/previous_chapters.py create mode 100644 Codes/ch07/01_main-chapter-code/tests.py create mode 100644 Codes/ch07/02_dataset-utilities/README.md create mode 100644 Codes/ch07/02_dataset-utilities/config.json create mode 100644 Codes/ch07/02_dataset-utilities/create-passive-voice-entries.ipynb create mode 100644 Codes/ch07/02_dataset-utilities/find-near-duplicates.py create mode 100644 Codes/ch07/02_dataset-utilities/instruction-examples-modified.json create mode 100644 Codes/ch07/02_dataset-utilities/instruction-examples.json create mode 100644 Codes/ch07/02_dataset-utilities/requirements-extra.txt create mode 100644 Codes/ch07/03_model-evaluation/README.md create mode 100644 Codes/ch07/03_model-evaluation/config.json create mode 100644 Codes/ch07/03_model-evaluation/eval-example-data.json create mode 100644 Codes/ch07/03_model-evaluation/llm-instruction-eval-ollama.ipynb create mode 100644 Codes/ch07/03_model-evaluation/llm-instruction-eval-openai.ipynb create mode 100644 Codes/ch07/03_model-evaluation/requirements-extra.txt create mode 100644 Codes/ch07/03_model-evaluation/scores/correlation-analysis.ipynb create mode 100644 Codes/ch07/03_model-evaluation/scores/gpt4-model-1-response.json create mode 100644 Codes/ch07/03_model-evaluation/scores/gpt4-model-2-response.json create mode 100644 Codes/ch07/03_model-evaluation/scores/llama3-8b-model-1-response.json create mode 100644 Codes/ch07/03_model-evaluation/scores/llama3-8b-model-2-response.json create mode 100644 Codes/ch07/04_preference-tuning-with-dpo/README.md create mode 100644 Codes/ch07/04_preference-tuning-with-dpo/create-preference-data-ollama.ipynb create mode 100644 Codes/ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb create mode 100644 Codes/ch07/04_preference-tuning-with-dpo/instruction-data-with-preference.json create mode 100644 Codes/ch07/04_preference-tuning-with-dpo/previous_chapters.py create mode 100644 Codes/ch07/05_dataset-generation/README.md create mode 100644 Codes/ch07/05_dataset-generation/instruction-data-llama3-7b.json create mode 100644 Codes/ch07/05_dataset-generation/llama3-ollama.ipynb create mode 100644 Codes/ch07/README.md diff --git a/Codes/ch07/01_main-chapter-code/README.md b/Codes/ch07/01_main-chapter-code/README.md new file mode 100644 index 0000000..9b9a877 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/README.md @@ -0,0 +1,76 @@ +# Chapter 7: Finetuning to Follow Instructions + +### Main Chapter Code + +- [ch07.ipynb](ch07.ipynb) contains all the code as it appears in the chapter +- [previous_chapters.py](previous_chapters.py) is a Python module that contains the GPT model we coded and trained in previous chapters, alongside many utility functions, which we reuse in this chapter +- [gpt_download.py](gpt_download.py) contains the utility functions for downloading the pretrained GPT model weights +- [exercise-solutions.ipynb](exercise-solutions.ipynb) contains the exercise solutions for this chapter + + +### Optional Code + +- [load-finetuned-model.ipynb](load-finetuned-model.ipynb) is a standalone Jupyter notebook to load the instruction finetuned model we created in this chapter + +- [gpt_instruction_finetuning.py](gpt_instruction_finetuning.py) is a standalone Python script to instruction finetune the model as described in the main chapter (think of it as a chapter summary focused on the finetuning parts) + +Usage: + +```bash +python gpt_instruction_finetuning.py +``` + +``` +matplotlib version: 3.9.0 +tiktoken version: 0.7.0 +torch version: 2.3.1 +tqdm version: 4.66.4 +tensorflow version: 2.16.1 +-------------------------------------------------- +Training set length: 935 +Validation set length: 55 +Test set length: 110 +-------------------------------------------------- +Device: cpu +-------------------------------------------------- +File already exists and is up-to-date: gpt2/355M/checkpoint +File already exists and is up-to-date: gpt2/355M/encoder.json +File already exists and is up-to-date: gpt2/355M/hparams.json +File already exists and is up-to-date: gpt2/355M/model.ckpt.data-00000-of-00001 +File already exists and is up-to-date: gpt2/355M/model.ckpt.index +File already exists and is up-to-date: gpt2/355M/model.ckpt.meta +File already exists and is up-to-date: gpt2/355M/vocab.bpe +Loaded model: gpt2-medium (355M) +-------------------------------------------------- +Initial losses + Training loss: 3.839039182662964 + Validation loss: 3.7619192123413088 +Ep 1 (Step 000000): Train loss 2.611, Val loss 2.668 +Ep 1 (Step 000005): Train loss 1.161, Val loss 1.131 +Ep 1 (Step 000010): Train loss 0.939, Val loss 0.973 +... +Training completed in 15.66 minutes. +Plot saved as loss-plot-standalone.pdf +-------------------------------------------------- +Generating responses +100%|█████████████████████████████████████████████████████████| 110/110 [06:57<00:00, 3.80s/it] +Responses saved as instruction-data-with-response-standalone.json +Model saved as gpt2-medium355M-sft-standalone.pth +``` + +- [ollama_evaluate.py](ollama_evaluate.py) is a standalone Python script to evaluate the responses of the finetuned model as described in the main chapter (think of it as a chapter summary focused on the evaluation parts) + +Usage: + +```bash +python ollama_evaluate.py --file_path instruction-data-with-response-standalone.json +``` + +``` +Ollama running: True +Scoring entries: 100%|███████████████████████████████████████| 110/110 [01:08<00:00, 1.62it/s] +Number of scores: 110 of 110 +Average score: 51.75 +``` + +- [exercise_experiments.py](exercise_experiments.py) is an optional scropt that implements the exercise solutions; for more details see [exercise-solutions.ipynb](exercise-solutions.ipynb) diff --git a/Codes/ch07/01_main-chapter-code/ch07.ipynb b/Codes/ch07/01_main-chapter-code/ch07.ipynb new file mode 100644 index 0000000..785e0ae --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/ch07.ipynb @@ -0,0 +1,2795 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "12e91914-5f51-43fa-b65b-625e73b4d17b", + "metadata": { + "id": "12e91914-5f51-43fa-b65b-625e73b4d17b" + }, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "c2520ec3-722f-4f44-bdd1-885b13e7afbf", + "metadata": { + "id": "c2520ec3-722f-4f44-bdd1-885b13e7afbf" + }, + "source": [ + "# 第7章：根据指令进行微调" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "4e19327b-6c02-4881-ad02-9b6d3ec0b1b4", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "4e19327b-6c02-4881-ad02-9b6d3ec0b1b4", + "outputId": "9d937b84-d8f8-4ce9-cc3c-211188f49a10" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "matplotlib version: 3.7.1\n", + "tiktoken version: 0.7.0\n", + "torch version: 2.4.0\n", + "tqdm version: 4.66.4\n", + "tensorflow version: 2.15.0\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\n", + " \"matplotlib\", # Plotting library\n", + " \"tiktoken\", # Tokenizer\n", + " \"torch\", # Deep learning library\n", + " \"tqdm\", # Progress bar\n", + " \"tensorflow\", # For OpenAI's pretrained weights\n", + "]\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "markdown", + "id": "264fca98-2f9a-4193-b435-2abfa3b4142f", + "metadata": { + "id": "264fca98-2f9a-4193-b435-2abfa3b4142f" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "8bbc68e9-75b3-41f1-ac2c-e071c3cd0813", + "metadata": { + "id": "8bbc68e9-75b3-41f1-ac2c-e071c3cd0813" + }, + "source": [ + "## 7.1 指令微调介绍" + ] + }, + { + "cell_type": "markdown", + "id": "53dba24a-6805-496c-9a7f-c75e2d3527ab", + "metadata": { + "id": "53dba24a-6805-496c-9a7f-c75e2d3527ab" + }, + "source": [ + "\n", + "- 在第5章中，我们看到预训练LLM涉及一个训练过程，该过程学习一次生成一个单词\n", + "- 因此，预训练的LLM擅长文本补全，但不擅长遵循指令\n", + "- 在本章中，我们将教LLM更好地遵循指令\n" + ] + }, + { + "cell_type": "markdown", + "id": "18dc0535-0904-44ed-beaf-9b678292ef35", + "metadata": { + "id": "18dc0535-0904-44ed-beaf-9b678292ef35" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "b4698b23-12e0-4bd7-a140-ccb3dd71d4e8", + "metadata": { + "id": "b4698b23-12e0-4bd7-a140-ccb3dd71d4e8" + }, + "source": [ + "\n", + "本章所涵盖的主题在下图中进行了总结\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "5384f0cf-ef3c-4436-a5fa-59bd25649f86", + "metadata": { + "id": "5384f0cf-ef3c-4436-a5fa-59bd25649f86" + }, + "source": [ + "## 7.2 为监督指令微调准备数据集" + ] + }, + { + "cell_type": "markdown", + "id": "f8b34ff8-619f-4e89-bd03-ce513269760d", + "metadata": { + "id": "f8b34ff8-619f-4e89-bd03-ce513269760d" + }, + "source": [ + "- 我们将使用我为这一章准备的一个指令数据集进行工作" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "0G3axLw6kY1N", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "0G3axLw6kY1N", + "outputId": "a5f70eb8-6248-4834-e7ae-6105e94e5afa" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of entries: 1100\n" + ] + } + ], + "source": [ + "import json\n", + "import os\n", + "import urllib\n", + "\n", + "\n", + "def download_and_load_file(file_path, url):\n", + "\n", + " if not os.path.exists(file_path):\n", + " with urllib.request.urlopen(url) as response:\n", + " text_data = response.read().decode(\"utf-8\")\n", + " with open(file_path, \"w\", encoding=\"utf-8\") as file:\n", + " file.write(text_data)\n", + " else:\n", + " with open(file_path, \"r\", encoding=\"utf-8\") as file:\n", + " text_data = file.read()\n", + "\n", + " with open(file_path, \"r\", encoding=\"utf-8\") as file:\n", + " data = json.load(file)\n", + "\n", + " return data\n", + "\n", + "\n", + "file_path = \"instruction-data.json\"\n", + "url = (\n", + " \"https://raw.githubusercontent.com/rasbt/LLMs-from-scratch\"\n", + " \"/main/ch07/01_main-chapter-code/instruction-data.json\"\n", + ")\n", + "\n", + "data = download_and_load_file(file_path, url)\n", + "print(\"Number of entries:\", len(data))" + ] + }, + { + "cell_type": "markdown", + "id": "d7af8176-4255-4e92-8c7d-998771733eb8", + "metadata": { + "id": "d7af8176-4255-4e92-8c7d-998771733eb8" + }, + "source": [ + "- 我们从上面的JSON文件中加载的`data`列表中的每个项目都是一个以下形式的字典" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "-LiuBMsHkzQV", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "-LiuBMsHkzQV", + "outputId": "cc742019-b8d7-40f9-b21a-6a5ddf821377" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Example entry:\n", + " {'instruction': 'Identify the correct spelling of the following word.', 'input': 'Ocassion', 'output': \"The correct spelling is 'Occasion.'\"}\n" + ] + } + ], + "source": [ + "print(\"Example entry:\\n\", data[50])" + ] + }, + { + "cell_type": "markdown", + "id": "c5a32b34-485a-4816-a77a-da14f9fe6e46", + "metadata": { + "id": "c5a32b34-485a-4816-a77a-da14f9fe6e46" + }, + "source": [ + "- Note that the `'input'` field can be empty:" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "uFInFxDDk2Je", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "uFInFxDDk2Je", + "outputId": "70241295-a9ec-4b7d-caf5-ab6f267e3271" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Another example entry:\n", + " {'instruction': \"What is an antonym of 'complicated'?\", 'input': '', 'output': \"An antonym of 'complicated' is 'simple'.\"}\n" + ] + } + ], + "source": [ + "print(\"Another example entry:\\n\", data[999])" + ] + }, + { + "cell_type": "markdown", + "id": "f034799a-6575-45fd-98c9-9d1012d0fd58", + "metadata": { + "id": "f034799a-6575-45fd-98c9-9d1012d0fd58" + }, + "source": [ + "\n", + "- 指令微调通常被称为“监督指令微调”，因为它涉及在数据集上训练模型，其中输入-输出对是明确提供的\n", + "- 将条目格式化为大型语言模型的输入有不同的方式；下面的图表展示了用于训练Alpaca的两个示例格式\n", + "(https://crfm.stanford.edu/2023/03/13/alpaca.html) 和 Phi-3 (https://arxiv.org/abs/2404.14219) LLMs, respectively" + ] + }, + { + "cell_type": "markdown", + "id": "dffa4f70-44d4-4be4-89a9-2159f4885b10", + "metadata": { + "id": "dffa4f70-44d4-4be4-89a9-2159f4885b10" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "dd79a74e-befb-491c-be49-f777a6a5b6a6", + "metadata": { + "id": "dd79a74e-befb-491c-be49-f777a6a5b6a6" + }, + "source": [ + "\n", + "- 在这一章中，我们使用了Alpaca风格的提示格式化，这是用于指令微调的原始提示模板。\n", + "- 下面，我们将格式化输入，作为输入传递给大型语言模型（LLM）。\n" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "Jhk37nnJnkBh", + "metadata": { + "id": "Jhk37nnJnkBh" + }, + "outputs": [], + "source": [ + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"Below is an instruction that describes a task. \"\n", + " f\"Write a response that appropriately completes the request.\"\n", + " f\"\\n\\n### Instruction:\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n\\n### Input:\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "markdown", + "id": "011e78b4-e89a-4653-a2ee-7b2739ca04d6", + "metadata": { + "id": "011e78b4-e89a-4653-a2ee-7b2739ca04d6" + }, + "source": [ + "- 带输入字段的格式化回复如下所示" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "F9UQRfjzo4Js", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "F9UQRfjzo4Js", + "outputId": "13ec7abf-ad94-4e26-860d-6a39a344f31f" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Identify the correct spelling of the following word.\n", + "\n", + "### Input:\n", + "Ocassion\n", + "\n", + "### Response:\n", + "The correct spelling is 'Occasion.'\n" + ] + } + ], + "source": [ + "model_input = format_input(data[50])\n", + "desired_response = f\"\\n\\n### Response:\\n{data[50]['output']}\"\n", + "\n", + "print(model_input + desired_response)" + ] + }, + { + "cell_type": "markdown", + "id": "4dc93ddf-431c-49c0-96f2-fb3a79c4d94c", + "metadata": { + "id": "4dc93ddf-431c-49c0-96f2-fb3a79c4d94c" + }, + "source": [ + "- 以下是格式化后的回复，但没有输入字段" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "a3891fa9-f738-41cd-946c-80ef9a99c346", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "a3891fa9-f738-41cd-946c-80ef9a99c346", + "outputId": "d6be5713-1293-4a70-c8c8-a86ea8e95817" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "What is an antonym of 'complicated'?\n", + "\n", + "### Response:\n", + "An antonym of 'complicated' is 'simple'.\n" + ] + } + ], + "source": [ + "model_input = format_input(data[999])\n", + "desired_response = f\"\\n\\n### Response:\\n{data[999]['output']}\"\n", + "\n", + "print(model_input + desired_response)" + ] + }, + { + "cell_type": "markdown", + "id": "4aa8afd5-2a21-49a5-90c3-6a03865a4771", + "metadata": { + "id": "4aa8afd5-2a21-49a5-90c3-6a03865a4771" + }, + "source": [ + "- 最后，在下一节中准备PyTorch数据加载器之前，我们将数据集分为训练集、验证集和测试集" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "aFZVopbIlNfx", + "metadata": { + "id": "aFZVopbIlNfx" + }, + "outputs": [], + "source": [ + "train_portion = int(len(data) * 0.85) # 85% for training\n", + "test_portion = int(len(data) * 0.1) # 10% for testing\n", + "val_portion = len(data) - train_portion - test_portion # Remaining 5% for validation\n", + "\n", + "train_data = data[:train_portion]\n", + "test_data = data[train_portion:train_portion + test_portion]\n", + "val_data = data[train_portion + test_portion:]" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "-zf6oht6bIUQ", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "-zf6oht6bIUQ", + "outputId": "bb5fe8e5-1ce5-4fca-a430-76ecf42e99ef" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Training set length: 935\n", + "Validation set length: 55\n", + "Test set length: 110\n" + ] + } + ], + "source": [ + "print(\"Training set length:\", len(train_data))\n", + "print(\"Validation set length:\", len(val_data))\n", + "print(\"Test set length:\", len(test_data))" + ] + }, + { + "cell_type": "markdown", + "id": "fcaaf606-f913-4445-8301-632ae10d387d", + "metadata": { + "id": "fcaaf606-f913-4445-8301-632ae10d387d" + }, + "source": [ + "## 7.3 将数据组织成训练批次" + ] + }, + { + "cell_type": "markdown", + "id": "233f63bd-9755-4d07-8884-5e2e5345cf27", + "metadata": { + "id": "233f63bd-9755-4d07-8884-5e2e5345cf27" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "c149fc1a-7757-4ec8-80cb-e2a3fb007a2c", + "metadata": { + "id": "c149fc1a-7757-4ec8-80cb-e2a3fb007a2c" + }, + "source": [ + "\n", + "- 我们分几个步骤来处理这个数据集批处理，如下图所示\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "b9af423f-aad9-4b3c-bea5-153021c04862", + "metadata": { + "id": "b9af423f-aad9-4b3c-bea5-153021c04862" + }, + "source": [ + "\n", + "- 首先，我们实现了一个`InstructionDataset`类，它在数据集中对所有输入进行了预分词，类似于第6章中的`SpamDataset`\n", + "\n", + "" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "adc29dc4-f1c7-4c71-937b-95119d6239bb", + "metadata": { + "id": "adc29dc4-f1c7-4c71-937b-95119d6239bb" + }, + "outputs": [], + "source": [ + "import torch\n", + "from torch.utils.data import Dataset\n", + "\n", + "\n", + "class InstructionDataset(Dataset):\n", + " def __init__(self, data, tokenizer):\n", + " self.data = data\n", + "\n", + " # Pre-tokenize texts\n", + " self.encoded_texts = []\n", + " for entry in data:\n", + " instruction_plus_input = format_input(entry)\n", + " response_text = f\"\\n\\n### Response:\\n{entry['output']}\"\n", + " full_text = instruction_plus_input + response_text\n", + " self.encoded_texts.append(\n", + " tokenizer.encode(full_text)\n", + " )\n", + "\n", + " def __getitem__(self, index):\n", + " return self.encoded_texts[index]\n", + "\n", + " def __len__(self):\n", + " return len(self.data)" + ] + }, + { + "cell_type": "markdown", + "id": "384f0e69-4b22-41c0-a25d-f077527eddd1", + "metadata": { + "id": "384f0e69-4b22-41c0-a25d-f077527eddd1" + }, + "source": [ + "\n", + "- 与第6章类似，我们希望在一个批次中收集多个训练示例以加速训练；这需要将所有输入填充到相似的长度\n", + "- 与前一章类似，我们使用`<|endoftext|>`标记作为填充标记\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "ff24fe1a-5746-461c-ad3d-b6d84a1a7c96", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "ff24fe1a-5746-461c-ad3d-b6d84a1a7c96", + "outputId": "4d63f8b8-b4ad-45d9-9e93-c9dd8c2b7706" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[50256]\n" + ] + } + ], + "source": [ + "import tiktoken\n", + "tokenizer = tiktoken.get_encoding(\"gpt2\")\n", + "\n", + "print(tokenizer.encode(\"<|endoftext|>\", allowed_special={\"<|endoftext|>\"}))" + ] + }, + { + "cell_type": "markdown", + "id": "9e5bd7bc-f347-4cf8-a0c2-94cb8799e427", + "metadata": { + "id": "9e5bd7bc-f347-4cf8-a0c2-94cb8799e427" + }, + "source": [ + "\n", + "- 在第6章中，我们将数据集中的所有示例填充到相同的长度\n", + " - 在这里，我们采用更复杂的方法，并开发了一个自定义的“合并”函数，可以将其传递给数据加载器\n", + " - 这个自定义的合并函数将每个批次中的训练示例填充到相同的长度（但不同的批次可以有不同的长度）\n" + ] + }, + { + "cell_type": "markdown", + "id": "65c4d943-4aa8-4a44-874e-05bc6831fbd3", + "metadata": { + "id": "65c4d943-4aa8-4a44-874e-05bc6831fbd3" + }, + "source": [ + "" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "eb4c77dd-c956-4a1b-897b-b466909f18ca", + "metadata": { + "id": "eb4c77dd-c956-4a1b-897b-b466909f18ca" + }, + "outputs": [], + "source": [ + "def custom_collate_draft_1(\n", + " batch,\n", + " pad_token_id=50256,\n", + " device=\"cpu\"\n", + "):\n", + " # Find the longest sequence in the batch\n", + " # and increase the max length by +1, which will add one extra\n", + " # padding token below\n", + " batch_max_length = max(len(item)+1 for item in batch)\n", + "\n", + " # Pad and prepare inputs\n", + " inputs_lst = []\n", + "\n", + " for item in batch:\n", + " new_item = item.copy()\n", + " # Add an <|endoftext|> token\n", + " new_item += [pad_token_id]\n", + " # Pad sequences to batch_max_length\n", + " padded = (\n", + " new_item + [pad_token_id] *\n", + " (batch_max_length - len(new_item))\n", + " )\n", + " # Via padded[:-1], we remove the extra padded token\n", + " # that has been added via the +1 setting in batch_max_length\n", + " # (the extra padding token will be relevant in later codes)\n", + " inputs = torch.tensor(padded[:-1])\n", + " inputs_lst.append(inputs)\n", + "\n", + " # Convert list of inputs to tensor and transfer to target device\n", + " # torch.stack的功能解释：\n", + " # 1)合并数据：在处理批量数据时，可以使用 torch.stack 将多个样本合并成一个批次。\n", + " # 2) 维度扩展：当需要在模型输入中增加维度时，可以使用此方法。 \n", + " inputs_tensor = torch.stack(inputs_lst).to(device)\n", + " return inputs_tensor" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "8fb02373-59b3-4f3a-b1d1-8181a2432645", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "8fb02373-59b3-4f3a-b1d1-8181a2432645", + "outputId": "8705ca9a-e999-4f70-9db8-1ad084eba7bb" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([[ 0, 1, 2, 3, 4],\n", + " [ 5, 6, 50256, 50256, 50256],\n", + " [ 7, 8, 9, 50256, 50256]])\n" + ] + } + ], + "source": [ + "inputs_1 = [0, 1, 2, 3, 4]\n", + "inputs_2 = [5, 6]\n", + "inputs_3 = [7, 8, 9]\n", + "\n", + "batch = (\n", + " inputs_1,\n", + " inputs_2,\n", + " inputs_3\n", + ")\n", + "\n", + "print(custom_collate_draft_1(batch))" + ] + }, + { + "cell_type": "markdown", + "id": "c46832ab-39b7-45f8-b330-ac9adfa10d1b", + "metadata": { + "id": "c46832ab-39b7-45f8-b330-ac9adfa10d1b" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "17769a19-b961-4213-92ef-34f441b2d1d6", + "metadata": { + "id": "17769a19-b961-4213-92ef-34f441b2d1d6" + }, + "source": [ + "\n", + "- 在上面，我们仅将输入返回给大型语言模型（LLM）；然而，对于LLM的训练，我们还需要目标值。\n", + "- 与预训练LLM类似，目标是输入向右移动1个位置后的值，因此LLM学习预测下一个标记（token）。\n" + ] + }, + { + "cell_type": "markdown", + "id": "0386b6fe-3455-4e70-becd-a5a4681ba2ef", + "metadata": { + "id": "0386b6fe-3455-4e70-becd-a5a4681ba2ef" + }, + "source": [ + "" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "74af192e-757c-4c0a-bdf9-b7eb25bf6ebc", + "metadata": { + "id": "74af192e-757c-4c0a-bdf9-b7eb25bf6ebc" + }, + "outputs": [], + "source": [ + "def custom_collate_draft_2(\n", + " batch,\n", + " pad_token_id=50256,\n", + " device=\"cpu\"\n", + "):\n", + " # Find the longest sequence in the batch\n", + " batch_max_length = max(len(item)+1 for item in batch)\n", + "\n", + " # Pad and prepare inputs\n", + " inputs_lst, targets_lst = [], []\n", + "\n", + " for item in batch:\n", + " new_item = item.copy()\n", + " # Add an <|endoftext|> token\n", + " new_item += [pad_token_id]\n", + " # Pad sequences to max_length\n", + " padded = (\n", + " new_item + [pad_token_id] *\n", + " (batch_max_length - len(new_item))\n", + " )\n", + " inputs = torch.tensor(padded[:-1]) # Truncate the last token for inputs\n", + " targets = torch.tensor(padded[1:]) # Shift +1 to the right for targets\n", + " inputs_lst.append(inputs)\n", + " targets_lst.append(targets)\n", + "\n", + " # Convert list of inputs to tensor and transfer to target device\n", + " inputs_tensor = torch.stack(inputs_lst).to(device)\n", + " targets_tensor = torch.stack(targets_lst).to(device)\n", + " return inputs_tensor, targets_tensor" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "6eb2bce3-28a7-4f39-9d4b-5e972d69066c", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "6eb2bce3-28a7-4f39-9d4b-5e972d69066c", + "outputId": "b9ceae14-13c2-49f7-f4a4-b503f3db3009" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([[ 0, 1, 2, 3, 4],\n", + " [ 5, 6, 50256, 50256, 50256],\n", + " [ 7, 8, 9, 50256, 50256]])\n", + "tensor([[ 1, 2, 3, 4, 50256],\n", + " [ 6, 50256, 50256, 50256, 50256],\n", + " [ 8, 9, 50256, 50256, 50256]])\n" + ] + } + ], + "source": [ + "inputs, targets = custom_collate_draft_2(batch)\n", + "print(inputs)\n", + "print(targets)" + ] + }, + { + "cell_type": "markdown", + "id": "3bf85703-a0e0-42aa-8f29-cbc28dbf4e15", + "metadata": { + "id": "3bf85703-a0e0-42aa-8f29-cbc28dbf4e15" + }, + "source": [ + "\n", + "- 接下来，我们引入一个`ignore_index`值来将所有填充令牌ID替换为一个新值；这个`ignore_index`的目的是我们可以在损失函数中忽略填充值（稍后会详细介绍）\n", + "\n", + "\n", + "\n", + "- 具体来说，这意味着我们将与`50256`相对应的标记ID替换为`-100`，如下所示" + ] + }, + { + "cell_type": "markdown", + "id": "bd4bed33-956e-4b3f-a09c-586d8203109a", + "metadata": { + "id": "bd4bed33-956e-4b3f-a09c-586d8203109a" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "5346513e-c3f4-44fe-af22-4ebd36497728", + "metadata": { + "id": "5346513e-c3f4-44fe-af22-4ebd36497728" + }, + "source": [ + "（此外，我们还引入了`allowed_max_length`，以便在需要限制样本长度时使用；如果您计划使用比GPT-2模型支持的1024个标记上下文大小更长的自己的数据集，这将非常有用）" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "41ec6e2d-9eb2-4124-913e-d2af39be4cf2", + "metadata": { + "id": "41ec6e2d-9eb2-4124-913e-d2af39be4cf2" + }, + "outputs": [], + "source": [ + "def custom_collate_fn(\n", + " batch,\n", + " pad_token_id=50256,\n", + " ignore_index=-100,\n", + " allowed_max_length=None,\n", + " device=\"cpu\"\n", + "):\n", + " # Find the longest sequence in the batch\n", + " batch_max_length = max(len(item)+1 for item in batch)\n", + "\n", + " # Pad and prepare inputs and targets\n", + " inputs_lst, targets_lst = [], []\n", + "\n", + " for item in batch:\n", + " new_item = item.copy()\n", + " # Add an <|endoftext|> token\n", + " new_item += [pad_token_id]\n", + " \n", + " # Pad sequences to max_length\n", + " padded = (\n", + " new_item + [pad_token_id] *\n", + " (batch_max_length - len(new_item))\n", + " )\n", + " inputs = torch.tensor(padded[:-1]) # Truncate the last token for inputs\n", + " targets = torch.tensor(padded[1:]) # Shift +1 to the right for targets\n", + "\n", + " # New: Replace all but the first padding tokens in targets by ignore_index\n", + " mask = targets == pad_token_id\n", + " indices = torch.nonzero(mask).squeeze()\n", + " if indices.numel() > 1:\n", + " targets[indices[1:]] = ignore_index\n", + "\n", + " # New: Optionally truncate to maximum sequence length\n", + " if allowed_max_length is not None:\n", + " inputs = inputs[:allowed_max_length]\n", + " targets = targets[:allowed_max_length]\n", + "\n", + " inputs_lst.append(inputs)\n", + " targets_lst.append(targets)\n", + "\n", + " # Convert list of inputs and targets to tensors and transfer to target device\n", + " inputs_tensor = torch.stack(inputs_lst).to(device)\n", + " targets_tensor = torch.stack(targets_lst).to(device)\n", + "\n", + " return inputs_tensor, targets_tensor" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "cdf5eec4-9ebe-4be0-9fca-9a47bee88fdc", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "cdf5eec4-9ebe-4be0-9fca-9a47bee88fdc", + "outputId": "a5501547-239d-431d-fb04-da7fa2ffad79" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([[ 0, 1, 2, 3, 4],\n", + " [ 5, 6, 50256, 50256, 50256],\n", + " [ 7, 8, 9, 50256, 50256]])\n", + "tensor([[ 1, 2, 3, 4, 50256],\n", + " [ 6, 50256, -100, -100, -100],\n", + " [ 8, 9, 50256, -100, -100]])\n" + ] + } + ], + "source": [ + "inputs, targets = custom_collate_fn(batch)\n", + "print(inputs)\n", + "print(targets)" + ] + }, + { + "cell_type": "markdown", + "id": "26727c90-0d42-43b3-af21-0a66ad4fbbc7", + "metadata": { + "id": "26727c90-0d42-43b3-af21-0a66ad4fbbc7" + }, + "source": [ + "\n", + "- 让我们看看用-100替换完成了什么\n", + "- 为了说明的目的，我们假设我们有一个小的分类任务，有两个类别标签，0和1，类似于第6章\n", + "- 如果我们有以下logits值（模型最后一层的输出），我们计算以下损失\n" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "W2jvh-OP9MFV", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "W2jvh-OP9MFV", + "outputId": "b5cd858e-7c58-4a21-c5a7-e72768bd301c" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(1.1269)\n" + ] + } + ], + "source": [ + "logits_1 = torch.tensor(\n", + " [[-1.0, 1.0], # 1st training example\n", + " [-0.5, 1.5]] # 2nd training example\n", + ")\n", + "targets_1 = torch.tensor([0, 1])\n", + "\n", + "\n", + "loss_1 = torch.nn.functional.cross_entropy(logits_1, targets_1)\n", + "print(loss_1)" + ] + }, + { + "cell_type": "markdown", + "id": "5edd3244-8886-4505-92e9-367d28529e1e", + "metadata": { + "id": "5edd3244-8886-4505-92e9-367d28529e1e" + }, + "source": [ + "- 现在，正如预期的那样，增加一个训练示例将影响损失" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "nvVMuil89v9N", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "nvVMuil89v9N", + "outputId": "e4a07b99-a23c-4404-ccdb-5f93c39f3b09" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(0.7936)\n" + ] + } + ], + "source": [ + "logits_2 = torch.tensor(\n", + " [[-1.0, 1.0],\n", + " [-0.5, 1.5],\n", + " [-0.5, 1.5]] # New 3rd training example\n", + ")\n", + "targets_2 = torch.tensor([0, 1, 1])\n", + "\n", + "loss_2 = torch.nn.functional.cross_entropy(logits_2, targets_2)\n", + "print(loss_2)" + ] + }, + { + "cell_type": "markdown", + "id": "54dca331-40e0-468b-b690-189fe156ba8f", + "metadata": { + "id": "54dca331-40e0-468b-b690-189fe156ba8f" + }, + "source": [ + "- 让我们看看如果我们把一个例子的类别标签替换为-100会发生什么" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "RTyB1vah9p56", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "RTyB1vah9p56", + "outputId": "28c16387-1d9c-48a7-eda7-aa270864683d" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(1.1269)\n", + "loss_1 == loss_3: tensor(True)\n" + ] + } + ], + "source": [ + "targets_3 = torch.tensor([0, 1, -100])\n", + "\n", + "loss_3 = torch.nn.functional.cross_entropy(logits_2, targets_3)\n", + "print(loss_3)\n", + "print(\"loss_1 == loss_3:\", loss_1 == loss_3)" + ] + }, + { + "cell_type": "markdown", + "id": "cef09d21-b652-4760-abea-4f76920e6a25", + "metadata": { + "id": "cef09d21-b652-4760-abea-4f76920e6a25" + }, + "source": [ + "- 如我们所见，这3个训练样本的结果损失与我们从2个训练样本计算得到的损失相同，这意味着交叉熵损失函数忽略了标签为-100的训练样本。\n", + "- 默认情况下，PyTorch具有cross_entropy(..., ignore_index=-100)设置，用于忽略与标签-100相对应的样本。\n", + "- 使用这个-100 ignore_index，我们可以忽略批次中用于将训练样本填充到相等长度的额外文本结束（填充）标记。\n", + "- 但是，我们不想忽略文本结束（填充）标记（50256）的第一个实例，因为它可以帮助大型语言模型（LLM）判断响应何时完成。" + ] + }, + { + "cell_type": "markdown", + "id": "6a4e9c5f-7c49-4321-9f1b-a50468a84524", + "metadata": { + "id": "6a4e9c5f-7c49-4321-9f1b-a50468a84524" + }, + "source": [ + "- 在实践中，如以下图所示，将对应指令的目标令牌ID屏蔽掉也是很常见的（这是完成本章后推荐给读者的一项练习）" + ] + }, + { + "cell_type": "markdown", + "id": "fab8f0ed-80e8-4fd9-bf84-e5d0e0bc0a39", + "metadata": { + "id": "fab8f0ed-80e8-4fd9-bf84-e5d0e0bc0a39" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "bccaf048-ec95-498c-9155-d5b3ccba6c96", + "metadata": { + "id": "bccaf048-ec95-498c-9155-d5b3ccba6c96" + }, + "source": [ + "## 7.4 为指令数据集创建数据加载器" + ] + }, + { + "cell_type": "markdown", + "id": "e6b8e656-3af3-4db6-8dde-d8c216a12f50", + "metadata": { + "id": "e6b8e656-3af3-4db6-8dde-d8c216a12f50" + }, + "source": [ + "在这一部分，我们使用`InstructionDataset`类和`custom_collate_fn`函数来实例化训练、验证和测试数据加载器" + ] + }, + { + "cell_type": "markdown", + "id": "9fffe390-b226-4d5c-983f-9f4da773cb82", + "metadata": { + "id": "9fffe390-b226-4d5c-983f-9f4da773cb82" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "932677e9-9317-42e8-b461-7b0269518f97", + "metadata": { + "id": "932677e9-9317-42e8-b461-7b0269518f97" + }, + "source": [ + "- 之前的custom_collate_fn函数的另一个额外细节是，我们现在直接将数据移动到目标设备（例如GPU），而不是在主训练循环中执行此操作，这提高了效率，因为当我们将custom_collate_fn作为数据加载器的一部分时，它可以作为后台进程执行\n", + "- 我们使用Python的functools标准库中的partial函数，通过预先填充原始函数的device参数来创建一个新函数" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "etpqqWh8phKc", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "etpqqWh8phKc", + "outputId": "925faf3a-6df4-4ad0-f276-f328493619c3" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Device: cuda\n" + ] + } + ], + "source": [ + "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", + "\n", + "# If you have a Mac with Apple Silicon chip, you can uncomment the next lines of code\n", + "# to train the model on the Mac's GPU cores. However, as of this writing, this results in\n", + "# larger numerical deviations from the results shown in this chapter, because Apple Silicon\n", + "# support in PyTorch is still experimental\n", + "\n", + "# if torch.backends.mps.is_available():\n", + "# device = torch.device(\"mps\")\n", + "\n", + "print(\"Device:\", device)" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "4e47fb30-c2c6-4e6d-a64c-76cc65be4a2c", + "metadata": { + "id": "4e47fb30-c2c6-4e6d-a64c-76cc65be4a2c" + }, + "outputs": [], + "source": [ + "from functools import partial\n", + "\n", + "customized_collate_fn = partial(\n", + " custom_collate_fn,\n", + " device=device,\n", + " allowed_max_length=1024\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "8ff42c29-8b81-45e5-ae8d-b97cd1cf447a", + "metadata": { + "id": "8ff42c29-8b81-45e5-ae8d-b97cd1cf447a" + }, + "source": [ + "- 接下来，我们实例化数据加载器，类似于前面的章节，但现在我们为批处理过程提供了自己的整理函数" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "BtWkgir6Hlpe", + "metadata": { + "id": "BtWkgir6Hlpe" + }, + "outputs": [], + "source": [ + "from torch.utils.data import DataLoader\n", + "\n", + "\n", + "num_workers = 0\n", + "batch_size = 8\n", + "\n", + "torch.manual_seed(123)\n", + "\n", + "train_dataset = InstructionDataset(train_data, tokenizer)\n", + "train_loader = DataLoader(\n", + " train_dataset,\n", + " batch_size=batch_size,\n", + " collate_fn=customized_collate_fn,\n", + " shuffle=True,\n", + " drop_last=True,\n", + " num_workers=num_workers\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "1d097dc8-ad34-4f05-b435-e4147965f532", + "metadata": { + "id": "1d097dc8-ad34-4f05-b435-e4147965f532" + }, + "outputs": [], + "source": [ + "val_dataset = InstructionDataset(val_data, tokenizer)\n", + "val_loader = DataLoader(\n", + " val_dataset,\n", + " batch_size=batch_size,\n", + " collate_fn=customized_collate_fn,\n", + " shuffle=False,\n", + " drop_last=False,\n", + " num_workers=num_workers\n", + ")\n", + "\n", + "test_dataset = InstructionDataset(test_data, tokenizer)\n", + "test_loader = DataLoader(\n", + " test_dataset,\n", + " batch_size=batch_size,\n", + " collate_fn=customized_collate_fn,\n", + " shuffle=False,\n", + " drop_last=False,\n", + " num_workers=num_workers\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "3f67c147-b1a2-4a95-9807-e2d0de0324c0", + "metadata": { + "id": "3f67c147-b1a2-4a95-9807-e2d0de0324c0" + }, + "source": [ + "- 让我们看看生成的输入和目标批次的维度是什么样的" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "GGs1AI3vHpnX", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "GGs1AI3vHpnX", + "outputId": "53a9695d-87cb-4d7c-8b43-1561dfa68ba0" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Train loader:\n", + "torch.Size([8, 61]) torch.Size([8, 61])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 73]) torch.Size([8, 73])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 72]) torch.Size([8, 72])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 62]) torch.Size([8, 62])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 62]) torch.Size([8, 62])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 77]) torch.Size([8, 77])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n", + "torch.Size([8, 79]) torch.Size([8, 79])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 83]) torch.Size([8, 83])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 60]) torch.Size([8, 60])\n", + "torch.Size([8, 59]) torch.Size([8, 59])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n", + "torch.Size([8, 63]) torch.Size([8, 63])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 91]) torch.Size([8, 91])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 89]) torch.Size([8, 89])\n", + "torch.Size([8, 59]) torch.Size([8, 59])\n", + "torch.Size([8, 88]) torch.Size([8, 88])\n", + "torch.Size([8, 83]) torch.Size([8, 83])\n", + "torch.Size([8, 83]) torch.Size([8, 83])\n", + "torch.Size([8, 70]) torch.Size([8, 70])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 74]) torch.Size([8, 74])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 83]) torch.Size([8, 83])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 60]) torch.Size([8, 60])\n", + "torch.Size([8, 60]) torch.Size([8, 60])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 61]) torch.Size([8, 61])\n", + "torch.Size([8, 58]) torch.Size([8, 58])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 63]) torch.Size([8, 63])\n", + "torch.Size([8, 87]) torch.Size([8, 87])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 61]) torch.Size([8, 61])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 60]) torch.Size([8, 60])\n", + "torch.Size([8, 72]) torch.Size([8, 72])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 70]) torch.Size([8, 70])\n", + "torch.Size([8, 57]) torch.Size([8, 57])\n", + "torch.Size([8, 72]) torch.Size([8, 72])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 62]) torch.Size([8, 62])\n", + "torch.Size([8, 74]) torch.Size([8, 74])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 70]) torch.Size([8, 70])\n", + "torch.Size([8, 91]) torch.Size([8, 91])\n", + "torch.Size([8, 61]) torch.Size([8, 61])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 81]) torch.Size([8, 81])\n", + "torch.Size([8, 74]) torch.Size([8, 74])\n", + "torch.Size([8, 82]) torch.Size([8, 82])\n", + "torch.Size([8, 63]) torch.Size([8, 63])\n", + "torch.Size([8, 83]) torch.Size([8, 83])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 77]) torch.Size([8, 77])\n", + "torch.Size([8, 91]) torch.Size([8, 91])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 61]) torch.Size([8, 61])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 78]) torch.Size([8, 78])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 64]) torch.Size([8, 64])\n", + "torch.Size([8, 83]) torch.Size([8, 83])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 74]) torch.Size([8, 74])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n" + ] + } + ], + "source": [ + "print(\"Train loader:\")\n", + "for inputs, targets in train_loader:\n", + " print(inputs.shape, targets.shape)" + ] + }, + { + "cell_type": "markdown", + "id": "0c8e8dd7-d46a-4cc3-8a7e-c1d31e1b4657", + "metadata": { + "id": "0c8e8dd7-d46a-4cc3-8a7e-c1d31e1b4657" + }, + "source": [ + "\n", + "- 如上所示，我们可以看到所有批次的大小均为8，但长度不同，符合预期\n", + "- 让我们也通过打印`inputs`批次中第一个训练示例的内容来双重检查输入是否包含对应于标记ID 50256的`<|endoftext|>`填充标记\n" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "21b8fd02-014f-4481-9b71-5bfee8f9dfcd", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "21b8fd02-014f-4481-9b71-5bfee8f9dfcd", + "outputId": "ce919ecd-5ded-453c-a312-10cf55c13da7" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([21106, 318, 281, 12064, 326, 8477, 257, 4876, 13, 19430,\n", + " 257, 2882, 326, 20431, 32543, 262, 2581, 13, 198, 198,\n", + " 21017, 46486, 25, 198, 30003, 6525, 262, 6827, 1262, 257,\n", + " 985, 576, 13, 198, 198, 21017, 23412, 25, 198, 464,\n", + " 5156, 318, 845, 13779, 13, 198, 198, 21017, 18261, 25,\n", + " 198, 464, 5156, 318, 355, 13779, 355, 257, 4936, 13,\n", + " 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256],\n", + " device='cuda:0')\n" + ] + } + ], + "source": [ + "print(inputs[0])" + ] + }, + { + "cell_type": "markdown", + "id": "5f1f3647-8971-4006-89e0-6a2a1ec1d360", + "metadata": { + "id": "5f1f3647-8971-4006-89e0-6a2a1ec1d360" + }, + "source": [ + "- 同样，我们视觉上双重检查目标是否包含-100占位符标记" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "id": "51649ab4-1a7e-4a9e-92c5-950a24fde211", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "51649ab4-1a7e-4a9e-92c5-950a24fde211", + "outputId": "fdf486f3-e99d-4891-9814-afc9e4991020", + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor([ 318, 281, 12064, 326, 8477, 257, 4876, 13, 19430, 257,\n", + " 2882, 326, 20431, 32543, 262, 2581, 13, 198, 198, 21017,\n", + " 46486, 25, 198, 30003, 6525, 262, 6827, 1262, 257, 985,\n", + " 576, 13, 198, 198, 21017, 23412, 25, 198, 464, 5156,\n", + " 318, 845, 13779, 13, 198, 198, 21017, 18261, 25, 198,\n", + " 464, 5156, 318, 355, 13779, 355, 257, 4936, 13, 50256,\n", + " -100, -100, -100, -100, -100, -100, -100, -100, -100],\n", + " device='cuda:0')\n" + ] + } + ], + "source": [ + "print(targets[0])" + ] + }, + { + "cell_type": "markdown", + "id": "d6aad445-8f19-4238-b9bf-db80767fb91a", + "metadata": { + "id": "d6aad445-8f19-4238-b9bf-db80767fb91a" + }, + "source": [ + "## 7.5 Loading a pretrained LLM" + ] + }, + { + "cell_type": "markdown", + "id": "5a5c07d1-4fc9-4846-94cf-b11a085a667b", + "metadata": { + "id": "5a5c07d1-4fc9-4846-94cf-b11a085a667b" + }, + "source": [ + "- 在本部分中，我们将使用与第5章5.5节和第6章6.4节中相同的代码来加载一个预训练的GPT模型。" + ] + }, + { + "cell_type": "markdown", + "id": "8d1b438f-88af-413f-96a9-f059c6c55fc4", + "metadata": { + "id": "8d1b438f-88af-413f-96a9-f059c6c55fc4" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "8c68eda7-e02e-4caa-846b-ca6dbd396ca2", + "metadata": { + "id": "8c68eda7-e02e-4caa-846b-ca6dbd396ca2" + }, + "source": [ + "-然而，我们并没有加载最小的1.24亿参数模型，而是加载了3.55亿参数的中等版本模型，因为1.24亿参数的模型太小，无法通过指令微调获得质量上合理的结果。" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "id": "0d249d67-5eba-414e-9bd2-972ebf01329d", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "0d249d67-5eba-414e-9bd2-972ebf01329d", + "outputId": "3f08f5e1-ca7c-406d-e2ae-1b5fcafad3f2" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-07-25 02:22:49.969483: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n", + "2024-07-25 02:22:50.023103: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n", + "2024-07-25 02:22:50.023136: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n", + "2024-07-25 02:22:50.024611: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n", + "2024-07-25 02:22:50.033304: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", + "To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", + "2024-07-25 02:22:51.282247: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n", + "checkpoint: 100%|██████████| 77.0/77.0 [00:00<00:00, 169kiB/s]\n", + "encoder.json: 100%|██████████| 1.04M/1.04M [00:00<00:00, 2.43MiB/s]\n", + "hparams.json: 100%|██████████| 91.0/91.0 [00:00<00:00, 168kiB/s]\n", + "model.ckpt.data-00000-of-00001: 100%|██████████| 1.42G/1.42G [00:56<00:00, 25.0MiB/s]\n", + "model.ckpt.index: 100%|██████████| 10.4k/10.4k [00:00<00:00, 16.5MiB/s]\n", + "model.ckpt.meta: 100%|██████████| 927k/927k [00:00<00:00, 1.96MiB/s]\n", + "vocab.bpe: 100%|██████████| 456k/456k [00:00<00:00, 1.53MiB/s]\n" + ] + } + ], + "source": [ + "from gpt_download import download_and_load_gpt2\n", + "from previous_chapters import GPTModel, load_weights_into_gpt\n", + "\n", + "\n", + "BASE_CONFIG = {\n", + " \"vocab_size\": 50257, # Vocabulary size\n", + " \"context_length\": 1024, # Context length\n", + " \"drop_rate\": 0.0, # Dropout rate\n", + " \"qkv_bias\": True # Query-key-value bias\n", + "}\n", + "\n", + "model_configs = {\n", + " \"gpt2-small (124M)\": {\"emb_dim\": 768, \"n_layers\": 12, \"n_heads\": 12},\n", + " \"gpt2-medium (355M)\": {\"emb_dim\": 1024, \"n_layers\": 24, \"n_heads\": 16},\n", + " \"gpt2-large (774M)\": {\"emb_dim\": 1280, \"n_layers\": 36, \"n_heads\": 20},\n", + " \"gpt2-xl (1558M)\": {\"emb_dim\": 1600, \"n_layers\": 48, \"n_heads\": 25},\n", + "}\n", + "\n", + "CHOOSE_MODEL = \"gpt2-medium (355M)\"\n", + "\n", + "BASE_CONFIG.update(model_configs[CHOOSE_MODEL])\n", + "\n", + "model_size = CHOOSE_MODEL.split(\" \")[-1].lstrip(\"(\").rstrip(\")\")\n", + "settings, params = download_and_load_gpt2(\n", + " model_size=model_size,\n", + " models_dir=\"gpt2\"\n", + ")\n", + "\n", + "model = GPTModel(BASE_CONFIG)\n", + "load_weights_into_gpt(model, params)\n", + "model.eval();" + ] + }, + { + "cell_type": "markdown", + "id": "dbf3afed-bc8e-4d3a-ad9d-eb6f57bb7af5", + "metadata": { + "id": "dbf3afed-bc8e-4d3a-ad9d-eb6f57bb7af5" + }, + "source": [ + "- 在我们开始对下一节中的模型进行微调之前，让我们看看它在其中一个验证任务上的表现如何" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "7bd32b7c-5b44-4d25-a09f-46836802ca74", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "7bd32b7c-5b44-4d25-a09f-46836802ca74", + "outputId": "30d4fbd9-7d22-4545-cfc5-c5749cc0bd93" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Convert the active sentence to passive: 'The chef cooks the meal every day.'\n" + ] + } + ], + "source": [ + "torch.manual_seed(123)\n", + "\n", + "input_text = format_input(val_data[0])\n", + "print(input_text)" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "2e3e68e0-2627-4c65-b4e7-1e0667e4f6fa", + "metadata": { + "id": "2e3e68e0-2627-4c65-b4e7-1e0667e4f6fa" + }, + "outputs": [], + "source": [ + "from previous_chapters import (\n", + " generate,\n", + " text_to_token_ids,\n", + " token_ids_to_text\n", + ")\n", + "\n", + "token_ids = generate(\n", + " model=model,\n", + " idx=text_to_token_ids(input_text, tokenizer),\n", + " max_new_tokens=35,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256,\n", + ")\n", + "generated_text = token_ids_to_text(token_ids, tokenizer)" + ] + }, + { + "cell_type": "markdown", + "id": "36e2fda5-f796-4954-8f72-1dd1123e3344", + "metadata": { + "id": "36e2fda5-f796-4954-8f72-1dd1123e3344" + }, + "source": [ + "\n", + "- 注意，我们在前几章中使用的 `generate` 函数返回的是输入和输出文本的合并结果，这在前面的部分中对于创建清晰的文本来说很方便。\n", + "- 为了分离出响应，我们可以从 `generated_text` 的开始处减去指令的长度。\n" + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "ba4a55bf-a245-48d8-beda-2838a58fb5ba", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "ba4a55bf-a245-48d8-beda-2838a58fb5ba", + "outputId": "b46de9b3-98f0-45e4-a9ae-86870c3244a1" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "The chef cooks the meal every day.\n", + "\n", + "### Instruction:\n", + "\n", + "Convert the active sentence to passive: 'The chef cooks the\n" + ] + } + ], + "source": [ + "response_text = (\n", + " generated_text[len(input_text):]\n", + " .replace(\"### Response:\", \"\")\n", + " .strip()\n", + ")\n", + "print(response_text)" + ] + }, + { + "cell_type": "markdown", + "id": "d44080b2-a4c5-4520-a797-549519f66a3e", + "metadata": { + "id": "d44080b2-a4c5-4520-a797-549519f66a3e" + }, + "source": [ + "- 如我们所见，该模型目前还不能按照指令操作；它创建了一个“回复”部分，但只是简单地重复了原始输入句子以及指令" + ] + }, + { + "cell_type": "markdown", + "id": "70d27b9d-a942-4cf5-b797-848c5f01e723", + "metadata": { + "id": "70d27b9d-a942-4cf5-b797-848c5f01e723" + }, + "source": [ + "## 7.6 Finetuning the LLM on instruction data" + ] + }, + { + "cell_type": "markdown", + "id": "314b2a39-88b4-44d8-8c85-1c5b0cd6cc4a", + "metadata": { + "id": "314b2a39-88b4-44d8-8c85-1c5b0cd6cc4a" + }, + "source": [ + "\n", + "在本节中，我们对模型进行微调\n", + "\n", + "\n", + "\n", + "- 注意，我们可以重用我们在前几章中使用的所有损失计算和训练函数" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "id": "65444865-df87-4d98-9faf-875e1c4be860", + "metadata": { + "id": "65444865-df87-4d98-9faf-875e1c4be860" + }, + "outputs": [], + "source": [ + "from previous_chapters import (\n", + " calc_loss_loader,\n", + " train_model_simple\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "00083059-aa41-4d37-8a17-1c72d1b1ca00", + "metadata": { + "id": "00083059-aa41-4d37-8a17-1c72d1b1ca00" + }, + "source": [ + "-在我们开始训练之前，我们先计算初始训练和验证集的损失（和之前的章节一样，目标是最小化损失）" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "d99fc6f8-63b2-43da-adbb-a7b6b92c8dd5", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "d99fc6f8-63b2-43da-adbb-a7b6b92c8dd5", + "outputId": "36fdf03b-6fa6-46c3-c77d-ecc99e886265" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Training loss: 3.82590970993042\n", + "Validation loss: 3.761933755874634\n" + ] + } + ], + "source": [ + "model.to(device)\n", + "\n", + "torch.manual_seed(123)\n", + "\n", + "with torch.no_grad():\n", + " train_loss = calc_loss_loader(train_loader, model, device, num_batches=5)\n", + " val_loss = calc_loss_loader(val_loader, model, device, num_batches=5)\n", + "\n", + "print(\"Training loss:\", train_loss)\n", + "print(\"Validation loss:\", val_loss)" + ] + }, + { + "cell_type": "markdown", + "id": "12a6da8f-15b3-42b0-a136-619b7a35c3e9", + "metadata": { + "id": "12a6da8f-15b3-42b0-a136-619b7a35c3e9" + }, + "source": [ + "- 请注意，由于我们使用的是较大的模型（3.55亿参数而不是1.24亿参数），因此训练成本略高于前几章。\n", + "- 下面给出了各种设备的运行时间作为参考（在兼容的GPU设备上运行此笔记本不需要对代码进行任何更改）" + ] + }, + { + "cell_type": "markdown", + "id": "db4b57fb-e689-4550-931c-6d34a932487c", + "metadata": { + "id": "db4b57fb-e689-4550-931c-6d34a932487c" + }, + "source": [ + "
\n", + " \n", + "| Model | Device | Runtime for 2 Epochs |\n", + "|--------------------|-----------------------|----------------------|\n", + "| gpt2-medium (355M) | CPU (M3 MacBook Air) | 15.78 minutes |\n", + "| gpt2-medium (355M) | GPU (M3 MacBook Air) | 10.77 minutes |\n", + "| gpt2-medium (355M) | GPU (L4) | 1.83 minutes |\n", + "| gpt2-medium (355M) | GPU (A100) | 0.86 minutes |\n", + "| gpt2-small (124M) | CPU (M3 MacBook Air) | 5.74 minutes |\n", + "| gpt2-small (124M) | GPU (M3 MacBook Air) | 3.73 minutes |\n", + "| gpt2-small (124M) | GPU (L4) | 0.69 minutes |\n", + "| gpt2-small (124M) | GPU (A100) | 0.39 minutes |\n", + "\n", + "
\n", + "\n", + "- 我使用 `\"gpt2-medium (355M)\"` 模型运行了这个笔记本" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "78bcf83a-1fff-4540-97c1-765c4016d5e3", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "78bcf83a-1fff-4540-97c1-765c4016d5e3", + "outputId": "cea0618c-56ca-418a-c972-bcc060362727" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Ep 1 (Step 000000): Train loss 2.637, Val loss 2.626\n", + "Ep 1 (Step 000005): Train loss 1.174, Val loss 1.102\n", + "Ep 1 (Step 000010): Train loss 0.872, Val loss 0.944\n", + "Ep 1 (Step 000015): Train loss 0.857, Val loss 0.906\n", + "Ep 1 (Step 000020): Train loss 0.776, Val loss 0.881\n", + "Ep 1 (Step 000025): Train loss 0.754, Val loss 0.859\n", + "Ep 1 (Step 000030): Train loss 0.799, Val loss 0.836\n", + "Ep 1 (Step 000035): Train loss 0.714, Val loss 0.808\n", + "Ep 1 (Step 000040): Train loss 0.672, Val loss 0.806\n", + "Ep 1 (Step 000045): Train loss 0.633, Val loss 0.789\n", + "Ep 1 (Step 000050): Train loss 0.663, Val loss 0.783\n", + "Ep 1 (Step 000055): Train loss 0.760, Val loss 0.763\n", + "Ep 1 (Step 000060): Train loss 0.719, Val loss 0.743\n", + "Ep 1 (Step 000065): Train loss 0.653, Val loss 0.735\n", + "Ep 1 (Step 000070): Train loss 0.532, Val loss 0.729\n", + "Ep 1 (Step 000075): Train loss 0.569, Val loss 0.728\n", + "Ep 1 (Step 000080): Train loss 0.605, Val loss 0.725\n", + "Ep 1 (Step 000085): Train loss 0.509, Val loss 0.709\n", + "Ep 1 (Step 000090): Train loss 0.562, Val loss 0.691\n", + "Ep 1 (Step 000095): Train loss 0.500, Val loss 0.681\n", + "Ep 1 (Step 000100): Train loss 0.503, Val loss 0.677\n", + "Ep 1 (Step 000105): Train loss 0.564, Val loss 0.670\n", + "Ep 1 (Step 000110): Train loss 0.555, Val loss 0.666\n", + "Ep 1 (Step 000115): Train loss 0.508, Val loss 0.664\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request. ### Instruction: Convert the active sentence to passive: 'The chef cooks the meal every day.' ### Response: The meal is prepared every day by the chef.<|endoftext|>The following is an instruction that describes a task. Write a response that appropriately completes the request. ### Instruction: Convert the active sentence to passive:\n", + "Ep 2 (Step 000120): Train loss 0.435, Val loss 0.672\n", + "Ep 2 (Step 000125): Train loss 0.451, Val loss 0.687\n", + "Ep 2 (Step 000130): Train loss 0.447, Val loss 0.683\n", + "Ep 2 (Step 000135): Train loss 0.405, Val loss 0.682\n", + "Ep 2 (Step 000140): Train loss 0.409, Val loss 0.681\n", + "Ep 2 (Step 000145): Train loss 0.369, Val loss 0.680\n", + "Ep 2 (Step 000150): Train loss 0.382, Val loss 0.675\n", + "Ep 2 (Step 000155): Train loss 0.413, Val loss 0.675\n", + "Ep 2 (Step 000160): Train loss 0.415, Val loss 0.683\n", + "Ep 2 (Step 000165): Train loss 0.379, Val loss 0.686\n", + "Ep 2 (Step 000170): Train loss 0.323, Val loss 0.681\n", + "Ep 2 (Step 000175): Train loss 0.337, Val loss 0.669\n", + "Ep 2 (Step 000180): Train loss 0.392, Val loss 0.656\n", + "Ep 2 (Step 000185): Train loss 0.415, Val loss 0.657\n", + "Ep 2 (Step 000190): Train loss 0.340, Val loss 0.648\n", + "Ep 2 (Step 000195): Train loss 0.330, Val loss 0.634\n", + "Ep 2 (Step 000200): Train loss 0.310, Val loss 0.634\n", + "Ep 2 (Step 000205): Train loss 0.352, Val loss 0.630\n", + "Ep 2 (Step 000210): Train loss 0.367, Val loss 0.630\n", + "Ep 2 (Step 000215): Train loss 0.394, Val loss 0.635\n", + "Ep 2 (Step 000220): Train loss 0.299, Val loss 0.648\n", + "Ep 2 (Step 000225): Train loss 0.346, Val loss 0.661\n", + "Ep 2 (Step 000230): Train loss 0.292, Val loss 0.659\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request. ### Instruction: Convert the active sentence to passive: 'The chef cooks the meal every day.' ### Response: The meal is cooked every day by the chef.<|endoftext|>The following is an instruction that describes a task. Write a response that appropriately completes the request. ### Instruction: What is the capital of the United Kingdom\n", + "Training completed in 1.84 minutes.\n" + ] + } + ], + "source": [ + "import time\n", + "\n", + "start_time = time.time()\n", + "\n", + "torch.manual_seed(123)\n", + "\n", + "optimizer = torch.optim.AdamW(model.parameters(), lr=0.00005, weight_decay=0.1)\n", + "\n", + "num_epochs = 2\n", + "\n", + "train_losses, val_losses, tokens_seen = train_model_simple(\n", + " model, train_loader, val_loader, optimizer, device,\n", + " num_epochs=num_epochs, eval_freq=5, eval_iter=5,\n", + " start_context=format_input(val_data[0]), tokenizer=tokenizer\n", + ")\n", + "\n", + "end_time = time.time()\n", + "execution_time_minutes = (end_time - start_time) / 60\n", + "print(f\"Training completed in {execution_time_minutes:.2f} minutes.\")" + ] + }, + { + "cell_type": "markdown", + "id": "Ise3wGjlB-iq", + "metadata": { + "id": "Ise3wGjlB-iq" + }, + "source": [ + "\n", + "- 根据上面的输出，我们可以看到模型训练得很好，因为训练损失和验证损失值都在减少\n", + "- 此外，根据每个epoch后打印的响应文本，我们可以看到模型正确地遵循了指令，将输入句子“'The chef cooks the meal every day.'”转换为被动语态“'The meal is cooked every day by the chef.'”（我们将在后面的部分中正确格式化和评估响应）\n", + "- 最后，让我们看看训练和验证损失曲线" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "id": "4acd368b-1403-4807-a218-9102e35bfdbb", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 308 + }, + "id": "4acd368b-1403-4807-a218-9102e35bfdbb", + "outputId": "680da58a-9bd7-402d-ac95-470a4a29a6c4" + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from previous_chapters import plot_losses\n", + "\n", + "epochs_tensor = torch.linspace(0, num_epochs, len(train_losses))\n", + "plot_losses(epochs_tensor, tokens_seen, train_losses, val_losses)" + ] + }, + { + "cell_type": "markdown", + "id": "6777e0c4-d82c-46d8-84fb-1376c4f8bae0", + "metadata": { + "id": "6777e0c4-d82c-46d8-84fb-1376c4f8bae0" + }, + "source": [ + "- 如我们所见，在第一个周期的开始，损失急剧下降，这意味着模型开始迅速学习\n", + "- 我们可以看到，在大约1个训练周期时，出现了轻微的过拟合现象" + ] + }, + { + "cell_type": "markdown", + "id": "87b79a47-13f9-4d1f-87b1-3339bafaf2a3", + "metadata": { + "id": "87b79a47-13f9-4d1f-87b1-3339bafaf2a3" + }, + "source": [ + "## 7.7 提取和保存响应" + ] + }, + { + "cell_type": "markdown", + "id": "5a25cc88-1758-4dd0-b8bf-c044cbf2dd49", + "metadata": { + "id": "5a25cc88-1758-4dd0-b8bf-c044cbf2dd49" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "17510e9d-7727-4d58-ba9a-d82ec23c1427", + "metadata": { + "id": "17510e9d-7727-4d58-ba9a-d82ec23c1427" + }, + "source": [ + "- 在这一节中，我们保存测试集响应以便在下一节中评分\n", + "- 我们还保存一份模型副本以备将来使用\n", + "- 但首先，让我们简要查看一下微调模型生成的响应" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "id": "VQ2NZMbfucAc", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "VQ2NZMbfucAc", + "outputId": "8416b4ac-1993-4628-dea6-7789cdc8926c" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Rewrite the sentence using a simile.\n", + "\n", + "### Input:\n", + "The car is very fast.\n", + "\n", + "Correct response:\n", + ">> The car is as fast as lightning.\n", + "\n", + "Model response:\n", + ">> The car is as fast as a bullet.\n", + "-------------------------------------\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "What type of cloud is typically associated with thunderstorms?\n", + "\n", + "Correct response:\n", + ">> The type of cloud typically associated with thunderstorms is cumulonimbus.\n", + "\n", + "Model response:\n", + ">> The type of cloud associated with thunderstorms is a cumulus cloud.\n", + "-------------------------------------\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Name the author of 'Pride and Prejudice'.\n", + "\n", + "Correct response:\n", + ">> Jane Austen.\n", + "\n", + "Model response:\n", + ">> The author of 'Pride and Prejudice' is Jane Austen.\n", + "-------------------------------------\n" + ] + } + ], + "source": [ + "torch.manual_seed(123)\n", + "\n", + "\n", + "for entry in test_data[:3]:\n", + "\n", + " input_text = format_input(entry)\n", + "\n", + " token_ids = generate(\n", + " model=model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + " response_text = (\n", + " generated_text[len(input_text):]\n", + " .replace(\"### Response:\", \"\")\n", + " .strip()\n", + ")\n", + "\n", + " print(input_text)\n", + " print(f\"\\nCorrect response:\\n>> {entry['output']}\")\n", + " print(f\"\\nModel response:\\n>> {response_text.strip()}\")\n", + " print(\"-------------------------------------\")" + ] + }, + { + "cell_type": "markdown", + "id": "49ab64c1-586f-4939-8def-23feeb1b3599", + "metadata": { + "id": "49ab64c1-586f-4939-8def-23feeb1b3599" + }, + "source": [ + "\n", + "- 根据测试集说明、给定响应和模型的响应，我们可以看出该模型表现相对较好\n", + "- 第一个和最后一个说明的答案显然是正确的\n", + "- 第二个答案很接近；模型回答的是“积云”而不是“积雨云”（但请注意，积云可以发展成积雨云，积雨云能够产生雷暴）\n", + "- 最重要的是，我们可以看到，模型评估并不像前一章那样直接，在那里我们只需要计算正确的垃圾邮件/非垃圾邮件类别标签的百分比来获得分类准确率\n", + "- 在实践中，像聊天机器人这样的指令微调大型语言模型（LLMs）通过多种方法进行评估\n", + " MMLU等短答案和多项选择基准测试 (\"Measuring Massive Multitask Language Understanding\", [https://arxiv.org/abs/2009.03300](https://arxiv.org/abs/2009.03300)),测试模型知识的测试\n", + " - 与其他大型语言模型（如LMSYS聊天机器人竞技场）相比的人类偏好比较 ([https://arena.lmsys.org](https://arena.lmsys.org))\n", + " - 自动对话基准测试，其中使用另一个像GPT-4这样的LLM来评估响应，例如AlpacaEval ([https://tatsu-lab.github.io/alpaca_eval/](https://tatsu-lab.github.io/alpaca_eval/))\n", + "- 在下一部分，我们将使用类似于AlpacaEval的方法，并使用另一个大型语言模型（LLM）来评估我们模型的响应；但是，我们将使用我们自己的测试集，而不是使用公开可用的基准数据集\n", + "- 为此，我们将模型响应添加到test_data字典中，并将其保存为\"instruction-data-with-response.json\"文件以进行记录，以便在需要时我们可以在单独的Python会话中加载和分析它" + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "id": "-PNGKzY4snKP", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "-PNGKzY4snKP", + "outputId": "0453dfb3-51cd-49e2-9e63-f65b606c3478" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|██████████| 110/110 [01:11<00:00, 1.54it/s]\n" + ] + } + ], + "source": [ + "from tqdm import tqdm\n", + "\n", + "for i, entry in tqdm(enumerate(test_data), total=len(test_data)):\n", + "\n", + " input_text = format_input(entry)\n", + "\n", + " token_ids = generate(\n", + " model=model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + " response_text = generated_text[len(input_text):].replace(\"### Response:\", \"\").strip()\n", + "\n", + " test_data[i][\"model_response\"] = response_text\n", + "\n", + "\n", + "with open(\"instruction-data-with-response.json\", \"w\") as file:\n", + " json.dump(test_data, file, indent=4) # \"indent\" for pretty-printing" + ] + }, + { + "cell_type": "markdown", + "id": "228d6fa7-d162-44c3-bef1-4013c027b155", + "metadata": { + "id": "228d6fa7-d162-44c3-bef1-4013c027b155" + }, + "source": [ + "- 让我们再次检查其中一个条目，看看响应是否已正确添加到`test_data`字典中" + ] + }, + { + "cell_type": "code", + "execution_count": 38, + "id": "u-AvCCMTnPSE", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "u-AvCCMTnPSE", + "outputId": "ce3b2545-8990-4446-e44c-a945e0049c06" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{'instruction': 'Rewrite the sentence using a simile.', 'input': 'The car is very fast.', 'output': 'The car is as fast as lightning.', 'model_response': 'The car is as fast as a bullet.'}\n" + ] + } + ], + "source": [ + "print(test_data[0])" + ] + }, + { + "cell_type": "markdown", + "id": "c1b2f3f6-8569-405a-9db6-d47cba65608a", + "metadata": { + "id": "c1b2f3f6-8569-405a-9db6-d47cba65608a" + }, + "source": [ + "- 最后，我们也保存了模型，以便将来需要时重复使用" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "id": "8cBU0iHmVfOI", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "8cBU0iHmVfOI", + "outputId": "d6e7f226-9310-43f5-f31f-adc3a893a8e9", + "scrolled": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Model saved as gpt2-medium355M-sft.pth\n" + ] + } + ], + "source": [ + "import re\n", + "\n", + "\n", + "file_name = f\"{re.sub(r'[ ()]', '', CHOOSE_MODEL) }-sft.pth\"\n", + "torch.save(model.state_dict(), file_name)\n", + "print(f\"Model saved as {file_name}\")\n", + "\n", + "# Load model via\n", + "# model.load_state_dict(torch.load(\"gpt2-medium355M-sft.pth\"))" + ] + }, + { + "cell_type": "markdown", + "id": "obgoGI89dgPm", + "metadata": { + "id": "obgoGI89dgPm" + }, + "source": [ + "## 7.8 评估微调后的大型语言模型（LLM）" + ] + }, + { + "cell_type": "markdown", + "id": "805b9d30-7336-499f-abb5-4a21be3129f5", + "metadata": { + "id": "805b9d30-7336-499f-abb5-4a21be3129f5" + }, + "source": [ + "" + ] + }, + { + "cell_type": "markdown", + "id": "68d2b9d3-b6ff-4533-a89d-7b66079b4fd1", + "metadata": { + "id": "68d2b9d3-b6ff-4533-a89d-7b66079b4fd1" + }, + "source": [ + "- 在本节中，我们使用另一个更大的LLM自动评估微调后的LLM的响应\n", + "- 特别是，我们使用Meta AI的经过指令微调的80亿参数的Llama 3模型，该模型可以通过ollama在本地运行 ([https://ollama.com](https://ollama.com))\n", + "-（或者，如果您更喜欢通过OpenAI API使用功能更强大的LLM，如GPT-4，请参见） [llm-instruction-eval-openai.ipynb](../03_model-evaluation/llm-instruction-eval-openai.ipynb) notebook)" + ] + }, + { + "cell_type": "markdown", + "id": "ea427a30-36ba-44e3-bb1f-eb0d7008d6e9", + "metadata": { + "id": "ea427a30-36ba-44e3-bb1f-eb0d7008d6e9" + }, + "source": [ + "- Ollama是一个用于高效运行LLM（大型语言模型）的应用程序\n", + "- 它是llama.cpp的一个包装器 ([https://github.com/ggerganov/llama.cpp](https://github.com/ggerganov/llama.cpp)),它使用纯C/C++实现大型语言模型（LLMs），以最大化效率。请注意，这是一个使用LLMs生成文本（推理）的工具，而不是用于训练或微调LLMs的工具。\n", + "- 在运行以下代码之前，请通过访问以下链接安装ollama： [https://ollama.com](https://ollama.com) 并按照说明操作（例如，点击“下载”按钮并为您的操作系统下载ollama应用程序）" + ] + }, + { + "cell_type": "markdown", + "id": "747a2fc7-282d-47ec-a987-ed0a23ed6822", + "metadata": { + "id": "747a2fc7-282d-47ec-a987-ed0a23ed6822" + }, + "source": [ + "- 对于 macOS 和 Windows 用户，点击您下载的 ollama 应用程序；如果它提示您安装命令行用法，请说“是”\n", + "- Linux 用户可以使用 ollama 网站上提供的安装命令\n", + "- 通常，在我们可以从命令行使用 ollama 之前，我们必须启动 ollama 应用程序或在单独的终端中运行 ollama serve\n", + "\n", + "\n", + "\n", + "\n", + "- 在另一个终端中运行ollama应用程序或ollama serve，然后在命令行中执行以下命令来尝试使用具有80亿参数的Llama 3模型（该模型占用4.7 GB的存储空间，首次执行此命令时会自动下载）\n", + "\n", + "```bash\n", + "# 8B model\n", + "ollama run llama3\n", + "```\n", + "\n", + "\n", + "- 输出如下所示\n", + "\n", + "```\n", + "$ ollama run llama3\n", + "pulling manifest\n", + "pulling 6a0746a1ec1a... 100% ▕████████████████▏ 4.7 GB\n", + "pulling 4fa551d4f938... 100% ▕████████████████▏  12 KB\n", + "pulling 8ab4849b038c... 100% ▕████████████████▏  254 B\n", + "pulling 577073ffcc6c... 100% ▕████████████████▏  110 B\n", + "pulling 3f8eb4da87fa... 100% ▕████████████████▏  485 B\n", + "verifying sha256 digest\n", + "writing manifest\n", + "removing any unused layers\n", + "success\n", + "```\n", + "\n", + "\n", + "- 注意，“llama3”指的是经过微调、具有80亿参数的Llama 3模型。\n", + "\n", + "- 使用ollama与“llama3”模型（即80亿参数的模型）需要16GB的RAM；如果你的机器不支持，你可以尝试较小的模型，比如通过将`model = \"phi-3\"`设置为38亿参数的phi-3模型，这只需要8GB的RAM。\n", + "\n", + "- 或者，如果你的机器支持，你也可以使用更大的700亿参数的Llama 3模型，只需将“llama3”替换为“llama3:70b”即可。\n", + "\n", + "- 下载完成后，你将看到一个命令行提示符，允许你与模型进行聊天。\n", + "\n", + "- 尝试一个像“羊驼吃什么？”这样的提示，它应该返回一个类似以下的输出。\n", + "\n", + "\n", + "```\n", + ">>> What do llamas eat?\n", + "Llamas are ruminant animals, which means they have a four-chambered\n", + "stomach and eat plants that are high in fiber. In the wild, llamas\n", + "typically feed on:\n", + "1. Grasses: They love to graze on various types of grasses, including tall\n", + "grasses, wheat, oats, and barley.\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "7b7b341c-ba0e-40bb-a52c-cb328bbd1fe4", + "metadata": { + "id": "7b7b341c-ba0e-40bb-a52c-cb328bbd1fe4" + }, + "source": [ + "- 您可以使用输入/bye来结束本次会话" + ] + }, + { + "cell_type": "markdown", + "id": "faaf3e02-8ca0-4edf-be23-60625a5b14e3", + "metadata": { + "id": "faaf3e02-8ca0-4edf-be23-60625a5b14e3" + }, + "source": [ + "- 以下代码在继续使用ollama评估我们在上一节中生成的测试集响应之前，会检查ollama会话是否正在正确运行" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "026e8570-071e-48a2-aa38-64d7be35f288", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 193 + }, + "id": "026e8570-071e-48a2-aa38-64d7be35f288", + "outputId": "e30d3533-e1f5-4aa9-b24f-33273fc7b30e" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Ollama running: True\n" + ] + } + ], + "source": [ + "import psutil\n", + "\n", + "def check_if_running(process_name):\n", + " running = False\n", + " for proc in psutil.process_iter([\"name\"]):\n", + " if process_name in proc.info[\"name\"]:\n", + " running = True\n", + " break\n", + " return running\n", + "\n", + "ollama_running = check_if_running(\"ollama\")\n", + "\n", + "if not ollama_running:\n", + " raise RuntimeError(\"Ollama not running. Launch ollama before proceeding.\")\n", + "print(\"Ollama running:\", check_if_running(\"ollama\"))" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "723c9b00-e3cd-4092-83c3-6e48b5cf65b0", + "metadata": { + "id": "723c9b00-e3cd-4092-83c3-6e48b5cf65b0" + }, + "outputs": [], + "source": [ + "# This cell is optional; it allows you to restart the notebook\n", + "# and only run section 7.7 without rerunning any of the previous code\n", + "import json\n", + "from tqdm import tqdm\n", + "\n", + "file_path = \"instruction-data-with-response.json\"\n", + "\n", + "with open(file_path, \"r\") as file:\n", + " test_data = json.load(file)\n", + "\n", + "\n", + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"Below is an instruction that describes a task. \"\n", + " f\"Write a response that appropriately completes the request.\"\n", + " f\"\\n\\n### Instruction:\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n\\n### Input:\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "markdown", + "id": "b3464705-d026-4594-977f-fb357e51c3a9", + "metadata": { + "id": "b3464705-d026-4594-977f-fb357e51c3a9" + }, + "source": [ + "- 现在，与之前我们用来与模型交互的`ollama run`命令不同，另一种方式是通过Python中的REST API使用以下函数进行交互\n", + "- 在您运行此笔记本中的下一个单元格之前，请确保ollama仍在运行（之前的代码单元格应打印出`\"Ollama running: True\"`）\n", + "- 接下来，运行以下代码单元格以查询模型" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "e3ae0e10-2b28-42ce-8ea2-d9366a58088f", + "metadata": { + "id": "e3ae0e10-2b28-42ce-8ea2-d9366a58088f" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Llamas are herbivores, which means they primarily feed on plant-based foods. Their diet typically consists of:\n", + "\n", + "1. Grasses: Llamas love to graze on various types of grasses, including tall grasses, short grasses, and even weeds.\n", + "2. Hay: High-quality hay, such as alfalfa or timothy hay, is a staple in a llama's diet. They enjoy the sweet taste and texture of fresh hay.\n", + "3. Grains: Llamas may receive grains like oats, barley, or corn as part of their daily ration. However, it's essential to provide these grains in moderation, as they can be high in calories.\n", + "4. Fruits and vegetables: Llamas enjoy a variety of fruits and veggies, such as apples, carrots, sweet potatoes, and leafy greens like kale or spinach.\n", + "5. Minerals: Llamas require access to mineral supplements, which help maintain their overall health and well-being.\n", + "\n", + "In the wild, llamas might also eat:\n", + "\n", + "1. Leaves: They'll munch on leaves from trees and shrubs, including plants like willow, alder, and birch.\n", + "2. Bark: In some cases, llamas may eat the bark of certain trees, like aspen or cottonwood.\n", + "3. Mosses and lichens: These non-vascular plants can be a tasty snack for llamas.\n", + "\n", + "In captivity, llama owners typically provide a balanced diet that includes a mix of hay, grains, and fruits/vegetables. It's essential to consult with a veterinarian or experienced llama breeder to determine the best feeding plan for your llama.\n" + ] + } + ], + "source": [ + "import urllib.request\n", + "\n", + "def query_model(\n", + " prompt,\n", + " model=\"llama3\",\n", + " url=\"http://localhost:11434/api/chat\"\n", + "):\n", + " # Create the data payload as a dictionary\n", + " data = {\n", + " \"model\": model,\n", + " \"messages\": [\n", + " {\"role\": \"user\", \"content\": prompt}\n", + " ],\n", + " \"options\": { # Settings below are required for deterministic responses\n", + " \"seed\": 123,\n", + " \"temperature\": 0,\n", + " \"num_ctx\": 2048\n", + " }\n", + " }\n", + "\n", + "\n", + " # Convert the dictionary to a JSON formatted string and encode it to bytes\n", + " payload = json.dumps(data).encode(\"utf-8\")\n", + "\n", + " # Create a request object, setting the method to POST and adding necessary headers\n", + " request = urllib.request.Request(\n", + " url,\n", + " data=payload,\n", + " method=\"POST\"\n", + " )\n", + " request.add_header(\"Content-Type\", \"application/json\")\n", + "\n", + " # Send the request and capture the response\n", + " response_data = \"\"\n", + " with urllib.request.urlopen(request) as response:\n", + " # Read and decode the response\n", + " while True:\n", + " line = response.readline().decode(\"utf-8\")\n", + " if not line:\n", + " break\n", + " response_json = json.loads(line)\n", + " response_data += response_json[\"message\"][\"content\"]\n", + "\n", + " return response_data\n", + "\n", + "\n", + "model = \"llama3\"\n", + "result = query_model(\"What do Llamas eat?\", model)\n", + "print(result)" + ] + }, + { + "cell_type": "markdown", + "id": "207ae28f-0f8c-4fda-aeef-e7e3046249cc", + "metadata": { + "id": "207ae28f-0f8c-4fda-aeef-e7e3046249cc" + }, + "source": [ + "- 现在，使用我们上面定义的`query_model`函数，我们可以评估我们的微调模型的响应；让我们在之前部分中查看的前3个测试集响应上尝试一下" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "86b839d4-064d-4178-b2d7-01691b452e5e", + "metadata": { + "id": "86b839d4-064d-4178-b2d7-01691b452e5e" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Dataset response:\n", + ">> The car is as fast as lightning.\n", + "\n", + "Model response:\n", + ">> The car is as fast as a bullet.\n", + "\n", + "Score:\n", + ">> I'd rate the model response \"The car is as fast as a bullet.\" an 85 out of 100.\n", + "\n", + "Here's why:\n", + "\n", + "* The response uses a simile correctly, comparing the speed of the car to something else (in this case, a bullet).\n", + "* The comparison is relevant and makes sense, as bullets are known for their high velocity.\n", + "* The phrase \"as fast as\" is used correctly to introduce the simile.\n", + "\n", + "The only reason I wouldn't give it a perfect score is that some people might find the comparison slightly less vivid or evocative than others. For example, comparing something to lightning (as in the original response) can be more dramatic and attention-grabbing. However, \"as fast as a bullet\" is still a strong and effective simile that effectively conveys the idea of the car's speed.\n", + "\n", + "Overall, I think the model did a great job!\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> The type of cloud typically associated with thunderstorms is cumulonimbus.\n", + "\n", + "Model response:\n", + ">> The type of cloud associated with thunderstorms is a cumulus cloud.\n", + "\n", + "Score:\n", + ">> I'd score this model response as 40 out of 100.\n", + "\n", + "Here's why:\n", + "\n", + "* The model correctly identifies that thunderstorms are related to clouds (correctly identifying the type of phenomenon).\n", + "* However, it incorrectly specifies the type of cloud associated with thunderstorms. Cumulus clouds are not typically associated with thunderstorms; cumulonimbus clouds are.\n", + "* The response lacks precision and accuracy in its description.\n", + "\n", + "Overall, while the model attempts to address the instruction, it provides an incorrect answer, which is a significant error.\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> Jane Austen.\n", + "\n", + "Model response:\n", + ">> The author of 'Pride and Prejudice' is Jane Austen.\n", + "\n", + "Score:\n", + ">> I'd rate my own response as 95 out of 100. Here's why:\n", + "\n", + "* The response accurately answers the question by naming the author of 'Pride and Prejudice' as Jane Austen.\n", + "* The response is concise and clear, making it easy to understand.\n", + "* There are no grammatical errors or ambiguities that could lead to confusion.\n", + "\n", + "The only reason I wouldn't give myself a perfect score is that the response is slightly redundant - it's not necessary to rephrase the question in the answer. A more concise response would be simply \"Jane Austen.\"\n", + "\n", + "-------------------------\n" + ] + } + ], + "source": [ + "for entry in test_data[:3]:\n", + " prompt = (\n", + " f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"score the model response `{entry['model_response']}`\"\n", + " f\" on a scale from 0 to 100, where 100 is the best score. \"\n", + " )\n", + " print(\"\\nDataset response:\")\n", + " print(\">>\", entry['output'])\n", + " print(\"\\nModel response:\")\n", + " print(\">>\", entry[\"model_response\"])\n", + " print(\"\\nScore:\")\n", + " print(\">>\", query_model(prompt))\n", + " print(\"\\n-------------------------\")" + ] + }, + { + "cell_type": "markdown", + "id": "b114fd65-9cfb-45f6-ab74-8331da136bf3", + "metadata": { + "id": "b114fd65-9cfb-45f6-ab74-8331da136bf3" + }, + "source": [ + "- 如我们所见，Llama 3模型提供了一个合理的评估，并且如果模型不完全正确，它也会给出部分分数，这可以从“积云”答案中看出\n", + "- 请注意，之前的提示返回了非常详细的评估；我们可以调整提示以生成0到100之间的整数响应（其中100为最佳），以计算我们模型的平均分数\n", + "- 在M3 MacBook Air笔记本电脑上，测试集中110个条目的评估大约需要1分钟" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "9d7bca69-97c4-47a5-9aa0-32f116fa37eb", + "metadata": { + "id": "9d7bca69-97c4-47a5-9aa0-32f116fa37eb" + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Scoring entries: 100%|████████████████████████| 110/110 [01:10<00:00, 1.57it/s]" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of scores: 110 of 110\n", + "Average score: 50.32\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "\n" + ] + } + ], + "source": [ + "def generate_model_scores(json_data, json_key, model=\"llama3\"):\n", + " scores = []\n", + " for entry in tqdm(json_data, desc=\"Scoring entries\"):\n", + " prompt = (\n", + " f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"score the model response `{entry[json_key]}`\"\n", + " f\" on a scale from 0 to 100, where 100 is the best score. \"\n", + " f\"Respond with the integer number only.\"\n", + " )\n", + " score = query_model(prompt, model)\n", + " try:\n", + " scores.append(int(score))\n", + " except ValueError:\n", + " print(f\"Could not convert score: {score}\")\n", + " continue\n", + "\n", + " return scores\n", + "\n", + "\n", + "scores = generate_model_scores(test_data, \"model_response\")\n", + "print(f\"Number of scores: {len(scores)} of {len(test_data)}\")\n", + "print(f\"Average score: {sum(scores)/len(scores):.2f}\\n\")" + ] + }, + { + "cell_type": "markdown", + "id": "407f08d5-9ada-4301-9ebc-f0533c76d3f2", + "metadata": { + "id": "407f08d5-9ada-4301-9ebc-f0533c76d3f2" + }, + "source": [ + "- 我们的模型平均得分超过50分，我们可以将其作为参考点，将模型与其他模型进行比较，或者尝试其他可能改进模型的训练设置\n", + "- 请注意，ollama在跨操作系统时（截至本文撰写时）并非完全确定性的，因此您获得的数字可能与上面显示的数字略有不同" + ] + }, + { + "cell_type": "markdown", + "id": "6408768b-2784-44f1-b48e-aed0c1eb9b94", + "metadata": { + "id": "6408768b-2784-44f1-b48e-aed0c1eb9b94" + }, + "source": [ + "\n", + "供参考：\n", + "\n", + "- 原始Llama 3 8B基础模型得分为58.51\n", + "- Llama 3 8B指令模型得分为82.65\n" + ] + }, + { + "cell_type": "markdown", + "id": "412d7325-284a-446c-92a1-5aa8acc52dee", + "metadata": { + "id": "412d7325-284a-446c-92a1-5aa8acc52dee" + }, + "source": [ + "## 7.9 结论" + ] + }, + { + "cell_type": "markdown", + "id": "tIbNMluCDjVM", + "metadata": { + "id": "tIbNMluCDjVM" + }, + "source": [ + "### 7.9.1 下一步是什么\n", + "- 这标志着本书的最后一章\n", + "- 我们涵盖了大型语言模型（LLM）开发周期的主要步骤：实现LLM架构、对LLM进行预训练，以及对其进行微调\n", + "\n", + "\n", + "\n", + "\n", + "- 正如本章所述，在指令微调之后有时会进行偏好微调，这是一个可选步骤\n", + "- 偏好微调过程对于将模型定制为更好地与特定用户偏好保持一致特别有用；如果您对此感兴趣，请查看[../04_preference-tuning-with-dpo](../04_preference-tuning-with-dpo)文件夹\n", + "\n", + "\n", + "- 这个GitHub仓库还包含了大量你可能喜欢的附加奖励材料；更多信息，请参见 [Bonus Material](https://github.com/rasbt/LLMs-from-scratch?tab=readme-ov-file#bonus-material) 这个存储库的README页面上的部分\n", + "\n", + "### 7.9.2 在快速变化的领域保持最新状态\n", + "\n", + "- 本部分无代码\n", + "\n", + "### 7.9.3 最后的话\n", + "\n", + "- 我希望你喜欢这个从零开始实现大型语言模型（LLM）并编写预训练和微调功能的旅程\n", + "- 在我看来，从零开始实现一个大型语言模型（LLM）是理解LLM工作原理的最佳方式；我希望你通过这种方法获得了更好的理解\n", + "- 虽然这本书主要用于教学目的，但你可能对在现实世界的应用中使用不同且更强大的LLM感兴趣\n", + "- 对此，您可以考虑使用诸如axolotl之类的流行工具 ([https://github.com/OpenAccess-AI-Collective/axolotl](https://github.com/OpenAccess-AI-Collective/axolotl)) or LitGPT ([https://github.com/Lightning-AI/litgpt](https://github.com/Lightning-AI/litgpt)),我帮助开发的" + ] + }, + { + "cell_type": "markdown", + "id": "f9853e7f-a81a-4806-9728-be1690807185", + "metadata": { + "id": "f9853e7f-a81a-4806-9728-be1690807185" + }, + "source": [ + "\n", + "## 摘要和要点\n", + "\n", + "- 请查看[./gpt_instruction_finetuning.py](./gpt_instruction_finetuning.py)脚本，这是一个用于分类微调的自包含脚本\n", + "- [./ollama_evaluate.py](./ollama_evaluate.py)是一个基于第7.8节的独立脚本，它使用Ollama和Llama 3评估包含“output”和“response”键的JSON文件\n", + "- [./load-finetuned-model.ipynb](./load-finetuned-model.ipynb)笔记本说明了如何在新会话中加载微调后的模型\n", + "- 您可以在[./exercise-solutions.ipynb](./exercise-solutions.ipynb)中找到练习解决方案\n" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "gpuType": "A100", + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.5" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/01_main-chapter-code/exercise-solutions.ipynb b/Codes/ch07/01_main-chapter-code/exercise-solutions.ipynb new file mode 100644 index 0000000..4533fc0 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/exercise-solutions.ipynb @@ -0,0 +1,1002 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "ba450fb1-8a26-4894-ab7a-5d7bfefe90ce", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "51c9672d-8d0c-470d-ac2d-1271f8ec3f14", + "metadata": {}, + "source": [ + "# Chapter 7 Exercise solutions" + ] + }, + { + "cell_type": "markdown", + "id": "2625ddc4-9cce-42bd-947d-4e2203fdc55c", + "metadata": {}, + "source": [ + "## Exercise 7.1: Changing prompt styles" + ] + }, + { + "cell_type": "markdown", + "id": "6be25a95-2a33-433b-a698-2365b5fc9357", + "metadata": {}, + "source": [ + "Suppose we have the following data entry:\n", + "\n", + "```json\n", + "{\n", + " \"instruction\": \"Identify the correct spelling of the following word.\",\n", + " \"input\": \"Ocassion\",\n", + " \"output\": \"The correct spelling is 'Occasion.'\"\n", + "}\n", + "```\n", + "\n", + "In the main chapter, we formatted it according to the Alpaca-style prompt template:\n", + "\n", + "```\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Identify the correct spelling of the following word.\n", + "\n", + "### Input:\n", + "Occassion\n", + "\n", + "### Response:\n", + "The correct spelling is 'Occasion.'\n", + "```\n", + "\n", + "In this exercise, we now use the Phi-3 prompt template instead, which formats the data entry as follows:\n", + "\n", + "```\n", + "\n", + "Identify the correct spelling of the following word: 'Occasion'\n", + "\n", + "\n", + "The correct spelling is 'Occasion'.\n", + "```\n", + "\n", + "Note that this prompt template is substantially shorter, which reduces the runtime and hardware requirements for finetuning the LLM and generating text since the input prompts are shorter.\n", + "To make this change, we update the `format_input` function as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "f99baa1e-c24c-417f-89d0-13e6d061ea6a", + "metadata": {}, + "outputs": [], + "source": [ + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"<|user|>\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "markdown", + "id": "e4ba538f-64b9-495d-847b-d9f1d324bc50", + "metadata": {}, + "source": [ + "Let's make sure that it works as intended by applying it to two input samples, one with and one without content in the `'input'` field:" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "877a57e2-535f-4363-b32a-a093edd951b8", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "<|user|>\n", + "Identify the correct spelling of the following word.\n", + "Ocassion\n", + "\n", + "<|user|>\n", + "What is an antonym of 'complicated'?\n" + ] + } + ], + "source": [ + "sample_data = [\n", + " {'instruction': 'Identify the correct spelling of the following word.', 'input': 'Ocassion', 'output': \"The correct spelling is 'Occasion.'\"}, \n", + " {'instruction': \"What is an antonym of 'complicated'?\", 'input': '', 'output': \"An antonym of 'complicated' is 'simple'.\"}\n", + "]\n", + "\n", + "print(format_input(sample_data[0]))\n", + "print()\n", + "print(format_input(sample_data[1]))" + ] + }, + { + "cell_type": "markdown", + "id": "fa2a6704-6c61-4a09-b8f5-ffc5a77d6aa3", + "metadata": {}, + "source": [ + "Next, we also update the `InstructionDataset` class to use the <|assistant|> prompt template for the response:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "17f1a42c-7cc0-4746-8a6d-3a4cb37e2ca1", + "metadata": {}, + "outputs": [], + "source": [ + "import tiktoken\n", + "from torch.utils.data import Dataset\n", + "\n", + "class InstructionDataset(Dataset):\n", + " def __init__(self, data, tokenizer):\n", + " self.data = data\n", + "\n", + " # Pre-tokenize texts\n", + " self.encoded_texts = []\n", + " for entry in data:\n", + "\n", + " ###################################################################\n", + " # NEW: Use `format_input_phi` and adjust the response text template\n", + " instruction_plus_input = format_input(entry)\n", + " response_text = f\"\\n<|assistant|>:\\n{entry['output']}\"\n", + " ###################################################################\n", + " full_text = instruction_plus_input + response_text\n", + " self.encoded_texts.append(\n", + " tokenizer.encode(full_text)\n", + " )\n", + "\n", + " def __getitem__(self, index):\n", + " return self.encoded_texts[index]\n", + "\n", + " def __len__(self):\n", + " return len(self.data)\n", + "\n", + "\n", + "tokenizer = tiktoken.get_encoding(\"gpt2\")" + ] + }, + { + "cell_type": "markdown", + "id": "e0650926-c39f-4442-8116-cb7494416f28", + "metadata": {}, + "source": [ + "Lastly, we also have to update the way we extract the generated response when we collect the test set responses:" + ] + }, + { + "cell_type": "markdown", + "id": "a9253041-812f-4a5f-9ab1-d7e4cb1407fb", + "metadata": {}, + "source": [ + "```python\n", + "for i, entry in tqdm(enumerate(test_data), total=len(test_data)):\n", + "\n", + " input_text = format_input(entry)\n", + " tokenizer=tokenizer\n", + "\n", + " token_ids = generate(\n", + " model=model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + "\n", + " # New: Adjust ###Response -> <|assistant|>\n", + " response_text = generated_text[len(input_text):].replace(\"<|assistant|>:\", \"\").strip()\n", + "\n", + " test_data[i][\"model_response\"] = response_text\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "29cd557c-3838-45e4-a26a-baed4b11175a", + "metadata": {}, + "source": [ + "For your convenience, the exercise solution is implemented in the [exercise_experiments.py](exercise_experiments.py) script, which you can run as follows:" + ] + }, + { + "cell_type": "markdown", + "id": "dd8158e9-cc70-4e0f-88b0-73c3e1d8c030", + "metadata": {}, + "source": [ + "```bash\n", + "python exercise_experiments.py --exercise_solution phi3_prompt\n", + "```\n", + "\n", + "Output:\n", + "\n", + "```\n", + "matplotlib version: 3.7.1\n", + "tiktoken version: 0.7.0\n", + "torch version: 2.3.0+cu121\n", + "tqdm version: 4.66.4\n", + "tensorflow version: 2.15.0\n", + "--------------------------------------------------\n", + "Training set length: 935\n", + "Validation set length: 55\n", + "Test set length: 110\n", + "--------------------------------------------------\n", + "Device: cuda\n", + "--------------------------------------------------\n", + "...\n", + "Loaded model: gpt2-medium (355M)\n", + "--------------------------------------------------\n", + "Initial losses\n", + " Training loss: 3.71630220413208\n", + " Validation loss: 3.6440994262695314\n", + "Ep 1 (Step 000000): Train loss 2.633, Val loss 2.622\n", + "...\n", + "Ep 2 (Step 000230): Train loss 0.424, Val loss 0.928\n", + "<|user|> Convert the active sentence to passive: 'The chef cooks the meal every day.' <|assistant|>: The meal is prepared every day by the chef....\n", + "Training completed in 1.50 minutes.\n", + "Plot saved as loss-plot-phi3-prompt.pdf\n", + "--------------------------------------------------\n", + "Generating responses\n", + "100% 110/110 [00:11<00:00, 9.27it/s]\n", + "Responses saved as instruction-data-with-response-phi3-prompt.json\n", + "Model saved as gpt2-medium355M-sft-phi3-prompt.pth\n", + "```\n", + "\n", + "For comparison, you can run the original chapter 7 finetuning code via `python exercise_experiments.py --exercise_solution baseline`. \n", + "\n", + "Note that on an Nvidia L4 GPU, the code above, using the Phi-3 prompt template, takes 1.5 min to run. In comparison, the Alpaca-style template takes 1.80 minutes to run. So, the Phi-3 template is approximately 17% faster since it results in shorter model inputs. \n", + "\n", + "Let's take a look at some of the responses to make sure they have been formatted correctly:\n", + "\n", + "```json\n", + " {\n", + " \"instruction\": \"Rewrite the sentence using a simile.\",\n", + " \"input\": \"The car is very fast.\",\n", + " \"output\": \"The car is as fast as lightning.\",\n", + " \"model_response\": \"The car is as fast as a cheetah.\"\n", + " },\n", + " {\n", + " \"instruction\": \"What type of cloud is typically associated with thunderstorms?\",\n", + " \"input\": \"\",\n", + " \"output\": \"The type of cloud typically associated with thunderstorms is cumulonimbus.\",\n", + " \"model_response\": \"The type of cloud associated with thunderstorms is a cumulus cloud.\"\n", + " },\n", + " {\n", + " \"instruction\": \"Name the author of 'Pride and Prejudice'.\",\n", + " \"input\": \"\",\n", + " \"output\": \"Jane Austen.\",\n", + " \"model_response\": \"The author of 'Pride and Prejudice' is Jane Austen.\"\n", + " },\n", + "```\n", + "\n", + "We can evaluate the performance using the Ollama Llama 3 method, which is for your convenience, also implemented in the `python exercise_experiments.py` script, which we can run as follows:\n", + "\n", + "```bash\n", + "python ollama_evaluate.py --file_path instruction-data-with-response-phi3-prompt.json\n", + "```\n", + "\n", + "Output:\n", + "\n", + "```\n", + "Ollama running: True\n", + "Scoring entries: 100%|████████████████████████| 110/110 [01:08<00:00, 1.60it/s]\n", + "Number of scores: 110 of 110\n", + "Average score: 48.87\n", + "```\n", + "\n", + "The score is close to 50, which is in the same ballpark as the score we previously achieved with the Alpaca-style prompts." + ] + }, + { + "cell_type": "markdown", + "id": "5fea8be3-30a1-4623-a6d7-b095c6c1092e", + "metadata": {}, + "source": [ + " \n", + "## Exercise 7.2: Instruction and input masking\n", + "\n", + "To mask out the instructions as shown in the following figure, we need to make slight modifications to the `InstructionDataset` class and `custom_collate_fn`.\n", + "\n", + "" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "4405196a-db81-470b-be39-167a059587b6", + "metadata": {}, + "outputs": [], + "source": [ + "# This `format_input` function is copied from the original chapter 7 code\n", + "\n", + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"Below is an instruction that describes a task. \"\n", + " f\"Write a response that appropriately completes the request.\"\n", + " f\"\\n\\n### Instruction:\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n\\n### Input:\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "markdown", + "id": "83658c09-af8a-425a-b940-eb1f06e43c0b", + "metadata": {}, + "source": [ + "We can modify the `InstructionDataset` class to collect the lengths of the instructions, which we will use in the collate function to locate the instruction content positions in the targets when we code the collate function, as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "e5e6188a-f182-4f26-b9e5-ccae3ecadae0", + "metadata": {}, + "outputs": [], + "source": [ + "import torch\n", + "from torch.utils.data import Dataset\n", + "\n", + "\n", + "class InstructionDataset(Dataset):\n", + " def __init__(self, data, tokenizer):\n", + " self.data = data\n", + "\n", + " ##########################################################################################\n", + " # New: Separate list for instruction lengths\n", + " self.instruction_lengths = []\n", + " ##########################################################################################\n", + " \n", + " self.encoded_texts = []\n", + " \n", + " for entry in data:\n", + " instruction_plus_input = format_input(entry)\n", + " response_text = f\"\\n\\n### Response:\\n{entry['output']}\"\n", + " full_text = instruction_plus_input + response_text\n", + " \n", + " self.encoded_texts.append(\n", + " tokenizer.encode(full_text)\n", + " )\n", + "\n", + " ##########################################################################################\n", + " # New: collect instruction lengths\n", + " instruction_length = len(tokenizer.encode(instruction_plus_input))\n", + " self.instruction_lengths.append(instruction_length)\n", + " ##########################################################################################\n", + " \n", + " def __getitem__(self, index):\n", + " # New: return both instruction lengths and texts separately\n", + " return self.instruction_lengths[index], self.encoded_texts[index]\n", + "\n", + " def __len__(self):\n", + " return len(self.data)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "0163b7d1-acb8-456c-8efe-86307b58f4bb", + "metadata": {}, + "outputs": [], + "source": [ + "import tiktoken\n", + "\n", + "tokenizer = tiktoken.get_encoding(\"gpt2\")" + ] + }, + { + "cell_type": "markdown", + "id": "3a186394-4960-424d-bb6a-f58459dd5994", + "metadata": {}, + "source": [ + "Next, we update the `custom_collate_fn` where each `batch` is now a tuple containing `(instruction_length, item)` instead of just `item` due to the changes in the `InstructionDataset` dataset. In addition, we now mask the corresponding instruction tokens in the target ID list." + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "f815e6fc-8e54-4105-aecd-d4c6e890ff9d", + "metadata": {}, + "outputs": [], + "source": [ + "def custom_collate_fn(\n", + " batch,\n", + " pad_token_id=50256,\n", + " ignore_index=-100,\n", + " allowed_max_length=None,\n", + " device=\"cpu\"\n", + "):\n", + " # Find the longest sequence in the batch\n", + " batch_max_length = max(len(item)+1 for instruction_length, item in batch) # New: batch is now a tuple\n", + "\n", + " # Pad and prepare inputs and targets\n", + " inputs_lst, targets_lst = [], []\n", + "\n", + " for instruction_length, item in batch: # New: batch is now a tuple\n", + " new_item = item.copy()\n", + " # Add an <|endoftext|> token\n", + " new_item += [pad_token_id]\n", + " # Pad sequences to max_length\n", + " padded = new_item + [pad_token_id] * (batch_max_length - len(new_item))\n", + " inputs = torch.tensor(padded[:-1]) # Truncate the last token for inputs\n", + " targets = torch.tensor(padded[1:]) # Shift +1 to the right for targets\n", + "\n", + " # Replace all but the first padding tokens in targets by ignore_index\n", + " mask = targets == pad_token_id\n", + " indices = torch.nonzero(mask).squeeze()\n", + " if indices.numel() > 1:\n", + " targets[indices[1:]] = ignore_index\n", + "\n", + " ##########################################################################################\n", + " # New: Mask all input and instruction tokens in the targets\n", + " targets[:instruction_length-1] = -100\n", + " ##########################################################################################\n", + " \n", + " # Optionally truncate to maximum sequence length\n", + " if allowed_max_length is not None:\n", + " inputs = inputs[:allowed_max_length]\n", + " targets = targets[:allowed_max_length]\n", + " \n", + " inputs_lst.append(inputs)\n", + " targets_lst.append(targets)\n", + "\n", + " # Convert list of inputs and targets to tensors and transfer to target device\n", + " inputs_tensor = torch.stack(inputs_lst).to(device)\n", + " targets_tensor = torch.stack(targets_lst).to(device)\n", + "\n", + " return inputs_tensor, targets_tensor" + ] + }, + { + "cell_type": "markdown", + "id": "0a4a4815-850e-42c4-b70d-67e8ce5ebd57", + "metadata": {}, + "source": [ + "Let's try it out on some sample data below:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "8da8a5b1-a8e2-4389-b21c-25b67be6dd1c", + "metadata": {}, + "outputs": [], + "source": [ + "sample_data = [\n", + " {'instruction': \"What is an antonym of 'complicated'?\", 'input': '', 'output': \"An antonym of 'complicated' is 'simple'.\"},\n", + " {'instruction': 'Sort the following list in alphabetical order.', 'input': 'Zebra, Elephant, Crocodile', 'output': 'Crocodile, Elephant, Zebra'},\n", + " {'instruction': 'Arrange the given numbers in descending order.', 'input': '5, 12, 8, 3, 15', 'output': '15, 12, 8, 5, 3.'}\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "435b0816-0fc8-4650-a84a-eceffa4d85e4", + "metadata": {}, + "outputs": [], + "source": [ + "from torch.utils.data import DataLoader\n", + "\n", + "train_dataset = InstructionDataset(sample_data, tokenizer)\n", + "train_loader = DataLoader(\n", + " train_dataset,\n", + " batch_size=len(sample_data),\n", + " collate_fn=custom_collate_fn,\n", + " num_workers=0\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "106bbbd7-7286-4eb6-b343-43419332a80f", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Train loader:\n", + "torch.Size([3, 64]) torch.Size([3, 64])\n" + ] + } + ], + "source": [ + "print(\"Train loader:\")\n", + "for inputs, targets in train_loader:\n", + " print(inputs.shape, targets.shape)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "9bb3288b-84a9-4962-ae59-a7a29fd34bce", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Inputs:\n", + " tensor([21106, 318, 281, 12064, 326, 8477, 257, 4876, 13, 19430,\n", + " 257, 2882, 326, 20431, 32543, 262, 2581, 13, 198, 198,\n", + " 21017, 46486, 25, 198, 42758, 262, 1708, 1351, 287, 24830,\n", + " 605, 1502, 13, 198, 198, 21017, 23412, 25, 198, 57,\n", + " 37052, 11, 42651, 11, 9325, 19815, 576, 198, 198, 21017,\n", + " 18261, 25, 198, 34, 12204, 375, 576, 11, 42651, 11,\n", + " 1168, 37052, 50256, 50256])\n", + "\n", + "\n", + "Targets:\n", + " tensor([ -100, -100, -100, -100, -100, -100, -100, -100, -100, -100,\n", + " -100, -100, -100, -100, -100, -100, -100, -100, -100, -100,\n", + " -100, -100, -100, -100, -100, -100, -100, -100, -100, -100,\n", + " -100, -100, -100, -100, -100, -100, -100, -100, -100, -100,\n", + " -100, -100, -100, -100, -100, -100, 198, 198, 21017, 18261,\n", + " 25, 198, 34, 12204, 375, 576, 11, 42651, 11, 1168,\n", + " 37052, 50256, -100, -100])\n" + ] + } + ], + "source": [ + "print(\"Inputs:\\n\", inputs[1])\n", + "print(\"\\n\\nTargets:\\n\", targets[1])" + ] + }, + { + "cell_type": "markdown", + "id": "cc40347b-2ca7-44e1-862d-0fd0c92f0628", + "metadata": {}, + "source": [ + "As we can see based on the `targets` tensor, both the instruction and padding tokens are now masked using the -100 placeholder tokens. \n", + "Let's decode the inputs just to make sure that they look correct:" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "76a9e6fa-3d75-4e39-b139-c3e05048f42b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Sort the following list in alphabetical order.\n", + "\n", + "### Input:\n", + "Zebra, Elephant, Crocodile\n", + "\n", + "### Response:\n", + "Crocodile, Elephant, Zebra<|endoftext|><|endoftext|>\n" + ] + } + ], + "source": [ + "print(tokenizer.decode(list(inputs[1])))" + ] + }, + { + "cell_type": "markdown", + "id": "845ebd36-f63f-4b58-a76e-7767e4d2ccbd", + "metadata": {}, + "source": [ + "Next, let's decode the non-masked target token IDS:" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "4d54a152-b778-455a-8941-e375e2a17e8f", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "\n", + "### Response:\n", + "Crocodile, Elephant, Zebra<|endoftext|>\n" + ] + } + ], + "source": [ + "non_masked_targets = targets[1][targets[1] != -100]\n", + "\n", + "print(tokenizer.decode(list(non_masked_targets)))" + ] + }, + { + "cell_type": "markdown", + "id": "3912bbf5-e9e2-474b-9552-d522e7510aa6", + "metadata": {}, + "source": [ + "As shown above, the non-masked target tokens exclude the `\"Instruction\"` and `\"Input\"` fields, as intended. Now, we can run the modified code to see how well the LLM performs when finetuned using this masking strategy.\n", + "\n", + "For your convenience, you can use the `exercise_experiments.py` code to run a comparison as follows:" + ] + }, + { + "cell_type": "markdown", + "id": "56a76097-9114-479d-8803-443b0ff48581", + "metadata": {}, + "source": [ + "```bash\n", + "python exercise_experiments.py --exercise_solution mask_instructions\n", + "```\n", + "\n", + "Output:\n", + "\n", + "```\n", + "matplotlib version: 3.7.1\n", + "tiktoken version: 0.7.0\n", + "torch version: 2.3.0+cu121\n", + "tqdm version: 4.66.4\n", + "tensorflow version: 2.15.0\n", + "--------------------------------------------------\n", + "Training set length: 935\n", + "Validation set length: 55\n", + "Test set length: 110\n", + "--------------------------------------------------\n", + "Device: cuda\n", + "--------------------------------------------------\n", + "...\n", + "Loaded model: gpt2-medium (355M)\n", + "--------------------------------------------------\n", + "Initial losses\n", + " Training loss: 2.280539035797119\n", + " Validation loss: 2.262560224533081\n", + "Ep 1 (Step 000000): Train loss 1.636, Val loss 1.620\n", + "...\n", + "Ep 2 (Step 000230): Train loss 0.143, Val loss 0.727\n", + "...\n", + "Training completed in 1.77 minutes.\n", + "Plot saved as loss-plot-mask-instructions.pdf\n", + "--------------------------------------------------\n", + "Generating responses\n", + "100% 110/110 [02:10<00:00, 1.19s/it]\n", + "Responses saved as instruction-data-with-response-mask-instructions.json\n", + "Model saved as gpt2-medium355M-sft-mask-instructions.pth\n", + "```\n", + "\n", + "Next, let's evaluate the performance of the resulting LLM:\n", + "\n", + "```bash\n", + "python ollama_evaluate.py --file_path instruction-data-with-response-mask-instructions.json\n", + "```\n", + "\n", + "```\n", + "Ollama running: True\n", + "Scoring entries: 100%|██████████████████████████████████████████████████████████████████████████████████████| 110/110 [01:23<00:00, 1.31it/s]\n", + "Number of scores: 110 of 110\n", + "Average score: 47.73\n", + "```\n", + "\n", + "As we can see based on the scores, the instruction masking does perform slightly worse, which is consistent with the observation in the \"Instruction Tuning With Loss Over Instructions\" paper (https://arxiv.org/abs/2405.14394)" + ] + }, + { + "cell_type": "markdown", + "id": "94a0f758-29da-44ee-b7af-32473b3c086e", + "metadata": {}, + "source": [ + " \n", + "## Exercise 7.3: Finetuning on the original Alpaca dataset" + ] + }, + { + "cell_type": "markdown", + "id": "68df7616-679f-4e53-954d-6e7cf2e2ef55", + "metadata": {}, + "source": [ + "To finetune the model on the original Stanford Alpaca dataset ([https://github.com/tatsu-lab/stanford_alpaca](https://github.com/tatsu-lab/stanford_alpaca)), you just need to change the file URL from\n", + "\n", + "```python\n", + "url = \"https://raw.githubusercontent.com/rasbt/LLMs-from-scratch/main/ch07/01_main-chapter-code/instruction-data.json\"\n", + "```\n", + "\n", + "to\n", + "\n", + "```python\n", + "url = \"https://raw.githubusercontent.com/tatsu-lab/stanford_alpaca/main/alpaca_data.json\"\n", + "```\n", + "\n", + "Note that the dataset contains 52k entries (50x more than in chapter 7), and the entries are longer than the ones we worked with in chapter 7.\n", + "Thus, it's highly recommended that the training be run on a GPU.\n", + "\n", + "If you encounter out-of-memory errors, consider reducing the batch size from 8 to 4, 2, or 1. In addition to lowering the batch size, you may also want to consider lowering the `allowed_max_length` from 1024 to 512 or 256." + ] + }, + { + "cell_type": "markdown", + "id": "d94c9621-2c3f-4551-b5b8-87cd96e38c9c", + "metadata": {}, + "source": [ + "For your convenience, you can use the `exercise_experiments.py` code to finetune the model on the 52k Alpaca dataset with a batch size of 4 and an `allowed_max_length` of 512 as follows:" + ] + }, + { + "cell_type": "markdown", + "id": "40a76486-73e6-4415-94dc-bfe2aa36ea52", + "metadata": {}, + "source": [ + "```bash\n", + "python exercise_experiments.py --exercise_solution alpaca_52k\n", + "```\n", + "\n", + "```\n", + "matplotlib version: 3.7.1\n", + "tiktoken version: 0.7.0\n", + "torch version: 2.3.0+cu121\n", + "tqdm version: 4.66.4\n", + "tensorflow version: 2.15.0\n", + "--------------------------------------------------\n", + "Training set length: 44201\n", + "Validation set length: 2601\n", + "Test set length: 5200\n", + "--------------------------------------------------\n", + "Device: cuda\n", + "--------------------------------------------------\n", + "...\n", + "Loaded model: gpt2-medium (355M)\n", + "--------------------------------------------------\n", + "Initial losses\n", + " Training loss: 3.3681655883789063\n", + " Validation loss: 3.4122894287109373\n", + "Ep 1 (Step 000000): Train loss 2.477, Val loss 2.750\n", + "...\n", + "Ep 2 (Step 022095): Train loss 0.761, Val loss 1.557\n", + "...\n", + "Training completed in 196.38 minutes.\n", + "Plot saved as loss-plot-alpaca52k.pdf\n", + "--------------------------------------------------\n", + "Generating responses\n", + "100% 5200/5200 [2:56:33<00:00, 2.04s/it]\n", + "Responses saved as instruction-data-with-response-alpaca52k.json\n", + "Model saved as gpt2-medium355M-sft-alpaca52k.pth\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "cc8a1bd1-15f4-4a35-87b5-369038db0aa7", + "metadata": {}, + "source": [ + "Below are a few examples from the Alpaca dataset, including the generated model responses:" + ] + }, + { + "cell_type": "markdown", + "id": "be9ab66f-5819-4b01-9a03-c45aa3b7c5b8", + "metadata": {}, + "source": [ + "```json\n", + "[\n", + " {\n", + " \"instruction\": \"Edit the following sentence to increase readability: \\\"He made a huge effort and was so successful.\\\"\",\n", + " \"input\": \"\",\n", + " \"output\": \"He exerted a tremendous effort, and thus enjoyed great success.\",\n", + " \"model_response\": \"He put in an immense effort and was rewarded with success.\"\n", + " },\n", + " {\n", + " \"instruction\": \"Rewrite the following sentence to make it more concise: \\\"I was displeased with the result of the experiment that I conducted.\\\"\",\n", + " \"input\": \"\",\n", + " \"output\": \"I was unhappy with my experiment's outcome.\",\n", + " \"model_response\": \"I was displeased with the results of the experiment.\"\n", + " },\n", + " {\n", + " \"instruction\": \"How can we build a more efficient GPT model?\",\n", + " \"input\": \"\",\n", + " \"output\": \"We can build a more efficient GPT model by optimizing the architecture of the model, using smaller model sizes and training with fewer parameters. We can also leverage techniques such as knowledge distillation, transfer learning, dynamic sparsity and hybrid computing to further improve the efficiency of the model.\",\n", + " \"model_response\": \"Building a more efficient GPT model requires careful planning and optimization. First, it is important to identify the target language and the context in which the model is used. Then, it is important to select the appropriate model architecture, such as backpropagation, hyperparameters, and hyperparameters. Finally, it is important to select the appropriate model weights and optimizers, such as backpropagation, hyperparameters, and hyperparameters.\"\n", + " },\n", + "]" + ] + }, + { + "cell_type": "markdown", + "id": "840e2076-f2e6-44a2-86fd-f191f9674267", + "metadata": {}, + "source": [ + "Finally, we can evaluate the finetuned LLM using the [ollama_evaluate.py](ollama_evaluate.py) utility function:\n", + "\n", + "```bash\n", + "python ollama_evaluate.py --file_path instruction-data-with-response-alpaca52k.json\n", + "```\n", + "\n", + "```\n", + "Scoring entries: 100%|████████████████████| 5200/5200 [1:07:52<00:00, 1.28it/s]\n", + "Number of scores: 5188 of 5200\n", + "Average score: 48.16\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "d14b3c60-00a1-43a9-9fcd-592aaadf1ef4", + "metadata": {}, + "source": [ + "The score is slightly lower than the score we obtained on the dataset we used in this chapter. However, note that the Alpaca test set contains more diverse and partly more challenging instructions than the dataset we used in the main chapter." + ] + }, + { + "cell_type": "markdown", + "id": "ca61fa6c-4e1d-4618-9e5e-d091f8303e30", + "metadata": {}, + "source": [ + "## Exercise 7.4: Parameter-efficient finetuning with LoRA" + ] + }, + { + "cell_type": "markdown", + "id": "01742cec-1f41-4415-8788-009d31b1ad38", + "metadata": {}, + "source": [ + "To instruction finetune the model using LoRA, use the relevant classes and functions from appendix E:\n", + "\n", + "```python\n", + "from appendix_E import LoRALayer, LinearWithLoRA, replace_linear_with_lora\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "871dca8f-3411-4735-b7b0-9d0e6e0599ac", + "metadata": {}, + "source": [ + "Next, add the following lines of code below the model loading code in section 7.5:\n", + "\n", + "\n", + "```python\n", + "total_params = sum(p.numel() for p in model.parameters() if p.requires_grad)\n", + "print(f\"Total trainable parameters before: {total_params:,}\")\n", + "\n", + "for param in model.parameters():\n", + " param.requires_grad = False\n", + "\n", + "total_params = sum(p.numel() for p in model.parameters() if p.requires_grad)\n", + "print(f\"Total trainable parameters after: {total_params:,}\")\n", + "replace_linear_with_lora(model, rank=16, alpha=16)\n", + "\n", + "total_params = sum(p.numel() for p in model.parameters() if p.requires_grad)\n", + "print(f\"Total trainable LoRA parameters: {total_params:,}\")\n", + "model.to(device)\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "1b26b925-dc95-4b91-b050-9676dd9608a4", + "metadata": {}, + "source": [ + "For your convenience, you can use the `exercise_experiments.py` code to finetune the model, using LoRA with rank 16 and alpa 16, as follows:" + ] + }, + { + "cell_type": "markdown", + "id": "01f02c7e-3b15-44b8-bf41-7892cd755766", + "metadata": {}, + "source": [ + "```bash\n", + "python exercise_experiments.py --exercise_solution lora\n", + "```\n", + "\n", + "Output:\n", + "\n", + "```\n", + "matplotlib version: 3.7.1\n", + "tiktoken version: 0.7.0\n", + "torch version: 2.3.0+cu121\n", + "tqdm version: 4.66.4\n", + "tensorflow version: 2.15.0\n", + "--------------------------------------------------\n", + "Training set length: 935\n", + "Validation set length: 55\n", + "Test set length: 110\n", + "--------------------------------------------------\n", + "Device: cuda\n", + "--------------------------------------------------\n", + "File already exists and is up-to-date: gpt2/355M/checkpoint\n", + "File already exists and is up-to-date: gpt2/355M/encoder.json\n", + "File already exists and is up-to-date: gpt2/355M/hparams.json\n", + "File already exists and is up-to-date: gpt2/355M/model.ckpt.data-00000-of-00001\n", + "File already exists and is up-to-date: gpt2/355M/model.ckpt.index\n", + "File already exists and is up-to-date: gpt2/355M/model.ckpt.meta\n", + "File already exists and is up-to-date: gpt2/355M/vocab.bpe\n", + "Loaded model: gpt2-medium (355M)\n", + "--------------------------------------------------\n", + "Total trainable parameters before: 406,286,336\n", + "Total trainable parameters after: 0\n", + "Total trainable LoRA parameters: 7,898,384\n", + "Initial losses\n", + " Training loss: 3.7684114456176756\n", + " Validation loss: 3.7619335651397705\n", + "Ep 1 (Step 000000): Train loss 2.509, Val loss 2.519\n", + "...\n", + "Ep 2 (Step 000230): Train loss 0.308, Val loss 0.652\n", + "...\n", + "--------------------------------------------------\n", + "Generating responses\n", + "100% 110/110 [01:52<00:00, 1.03s/it]\n", + "Responses saved as instruction-data-with-response-lora.json\n", + "Model saved as gpt2-medium355M-sft-lora.pth\n", + "```\n", + "\n", + "For comparison, you can run the original chapter 7 finetuning code via `python exercise_experiments.py --exercise_solution baseline`. \n", + "\n", + "Note that on an Nvidia L4 GPU, the code above, using LoRA, takes 1.30 min to run. In comparison, the baseline takes 1.80 minutes to run. So, LoRA is approximately 28% faster.\n", + "\n", + "\n", + "We can evaluate the performance using the Ollama Llama 3 method, which is for your convenience, also implemented in the `python exercise_experiments.py` script, which we can run as follows:\n", + "\n", + "```bash\n", + "python ollama_evaluate.py --file_path instruction-data-with-response-lora.json\n", + "```\n", + "\n", + "Output:\n", + "\n", + "```\n", + "Ollama running: True\n", + "Scoring entries: 100%|████████████████████████| 110/110 [01:13<00:00, 1.50it/s]\n", + "Number of scores: 110 of 110\n", + "Average score: 50.23\n", + "```\n", + "\n", + "The score is around 50, which is in the same ballpark as the original model." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.11" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/01_main-chapter-code/exercise_experiments.py b/Codes/ch07/01_main-chapter-code/exercise_experiments.py new file mode 100644 index 0000000..840284a --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/exercise_experiments.py @@ -0,0 +1,562 @@ +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch +# +# Code to run the exercises; see exercise-solutions.ipynb for more information + +from functools import partial +from importlib.metadata import version +import json +import math +import os +import re +import time +import urllib + +import matplotlib.pyplot as plt +from matplotlib.ticker import MaxNLocator +import tiktoken +import torch +from torch.utils.data import Dataset, DataLoader +from tqdm import tqdm + +# Import from local files in this folder +from gpt_download import download_and_load_gpt2 +from previous_chapters import ( + calc_loss_loader, + generate, + GPTModel, + load_weights_into_gpt, + text_to_token_ids, + train_model_simple, + token_ids_to_text +) + + +class InstructionDataset(Dataset): + def __init__(self, data, tokenizer): + self.data = data + + # Pre-tokenize texts + self.encoded_texts = [] + for entry in data: + instruction_plus_input = format_input(entry) + response_text = f"\n\n### Response:\n{entry['output']}" + full_text = instruction_plus_input + response_text + self.encoded_texts.append( + tokenizer.encode(full_text) + ) + + def __getitem__(self, index): + return self.encoded_texts[index] + + def __len__(self): + return len(self.data) + + +class InstructionDatasetWithMasking(Dataset): + def __init__(self, data, tokenizer): + self.data = data + + # New: Separate list for instruction lengths + self.instruction_lengths = [] + self.encoded_texts = [] + + for entry in data: + instruction_plus_input = format_input(entry) + response_text = f"\n\n### Response:\n{entry['output']}" + full_text = instruction_plus_input + response_text + + self.encoded_texts.append( + tokenizer.encode(full_text) + ) + + # New: collect instruction lengths + instruction_length = len(tokenizer.encode(instruction_plus_input)) + self.instruction_lengths.append(instruction_length) + + def __getitem__(self, index): + # New: return both instruction lengths and texts separately + return self.instruction_lengths[index], self.encoded_texts[index] + + def __len__(self): + return len(self.data) + + +class InstructionDatasetPhi(Dataset): + def __init__(self, data, tokenizer): + self.data = data + + # Pre-tokenize texts + self.encoded_texts = [] + for entry in data: + + ################################################################### + # NEW: Use `format_input_phi` and adjust the response text template + instruction_plus_input = format_input_phi(entry) + response_text = f"\n<|assistant|>:\n{entry['output']}" + ################################################################### + full_text = instruction_plus_input + response_text + self.encoded_texts.append( + tokenizer.encode(full_text) + ) + + def __getitem__(self, index): + return self.encoded_texts[index] + + def __len__(self): + return len(self.data) + + +class LinearWithLoRA(torch.nn.Module): + def __init__(self, linear, rank, alpha): + super().__init__() + self.linear = linear + self.lora = LoRALayer( + linear.in_features, linear.out_features, rank, alpha + ) + + def forward(self, x): + return self.linear(x) + self.lora(x) + + +class LoRALayer(torch.nn.Module): + def __init__(self, in_dim, out_dim, rank, alpha): + super().__init__() + self.A = torch.nn.Parameter(torch.empty(in_dim, rank)) + torch.nn.init.kaiming_uniform_(self.A, a=math.sqrt(5)) # similar to standard weight initialization + self.B = torch.nn.Parameter(torch.zeros(rank, out_dim)) + self.alpha = alpha + + def forward(self, x): + x = self.alpha * (x @ self.A @ self.B) + return x + + +def replace_linear_with_lora(model, rank, alpha): + for name, module in model.named_children(): + if isinstance(module, torch.nn.Linear): + # Replace the Linear layer with LinearWithLoRA + setattr(model, name, LinearWithLoRA(module, rank, alpha)) + else: + # Recursively apply the same function to child modules + replace_linear_with_lora(module, rank, alpha) + + +def custom_collate_fn( + batch, + pad_token_id=50256, + ignore_index=-100, + allowed_max_length=None, + device="cpu" +): + # Find the longest sequence in the batch + batch_max_length = max(len(item)+1 for item in batch) + + # Pad and prepare inputs and targets + inputs_lst, targets_lst = [], [] + + for item in batch: + new_item = item.copy() + # Add an <|endoftext|> token + new_item += [pad_token_id] + # Pad sequences to max_length + padded = new_item + [pad_token_id] * (batch_max_length - len(new_item)) + inputs = torch.tensor(padded[:-1]) # Truncate the last token for inputs + targets = torch.tensor(padded[1:]) # Shift +1 to the right for targets + + # New: Replace all but the first padding tokens in targets by ignore_index + mask = targets == pad_token_id + indices = torch.nonzero(mask).squeeze() + if indices.numel() > 1: + targets[indices[1:]] = ignore_index + + # New: Optionally truncate to maximum sequence length + if allowed_max_length is not None: + inputs = inputs[:allowed_max_length] + targets = targets[:allowed_max_length] + + inputs_lst.append(inputs) + targets_lst.append(targets) + + # Convert list of inputs and targets to tensors and transfer to target device + inputs_tensor = torch.stack(inputs_lst).to(device) + targets_tensor = torch.stack(targets_lst).to(device) + + return inputs_tensor, targets_tensor + + +def custom_collate_with_masking_fn( + batch, + pad_token_id=50256, + ignore_index=-100, + allowed_max_length=None, + device="cpu" +): + # Find the longest sequence in the batch + batch_max_length = max(len(item)+1 for instruction_length, item in batch) # New: batch is now a tuple + + # Pad and prepare inputs and targets + inputs_lst, targets_lst = [], [] + + for instruction_length, item in batch: # New: batch is now a tuple + new_item = item.copy() + # Add an <|endoftext|> token + new_item += [pad_token_id] + # Pad sequences to max_length + padded = new_item + [pad_token_id] * (batch_max_length - len(new_item)) + inputs = torch.tensor(padded[:-1]) # Truncate the last token for inputs + targets = torch.tensor(padded[1:]) # Shift +1 to the right for targets + + # Replace all but the first padding tokens in targets by ignore_index + mask = targets == pad_token_id + indices = torch.nonzero(mask).squeeze() + if indices.numel() > 1: + targets[indices[1:]] = ignore_index + + # New: Mask all input and instruction tokens in the targets + targets[:instruction_length-1] = -100 + + # Optionally truncate to maximum sequence length + if allowed_max_length is not None: + inputs = inputs[:allowed_max_length] + targets = targets[:allowed_max_length] + + inputs_lst.append(inputs) + targets_lst.append(targets) + + # Convert list of inputs and targets to tensors and transfer to target device + inputs_tensor = torch.stack(inputs_lst).to(device) + targets_tensor = torch.stack(targets_lst).to(device) + + return inputs_tensor, targets_tensor + + +def download_and_load_file(file_path, url): + + if not os.path.exists(file_path): + with urllib.request.urlopen(url) as response: + text_data = response.read().decode("utf-8") + with open(file_path, "w", encoding="utf-8") as file: + file.write(text_data) + else: + with open(file_path, "r", encoding="utf-8") as file: + text_data = file.read() + + with open(file_path, "r") as file: + data = json.load(file) + + return data + + +def format_input_phi(entry): + instruction_text = ( + f"<|user|>\n{entry['instruction']}" + ) + + input_text = f"\n{entry['input']}" if entry["input"] else "" + + return instruction_text + input_text + + +def format_input(entry): + instruction_text = ( + f"Below is an instruction that describes a task. " + f"Write a response that appropriately completes the request." + f"\n\n### Instruction:\n{entry['instruction']}" + ) + + input_text = f"\n\n### Input:\n{entry['input']}" if entry["input"] else "" + + return instruction_text + input_text + + +def plot_losses(epochs_seen, tokens_seen, train_losses, val_losses, plot_name): + fig, ax1 = plt.subplots(figsize=(12, 6)) + + # Plot training and validation loss against epochs + ax1.plot(epochs_seen, train_losses, label="Training loss") + ax1.plot(epochs_seen, val_losses, linestyle="-.", label="Validation loss") + ax1.set_xlabel("Epochs") + ax1.set_ylabel("Loss") + ax1.legend(loc="upper right") + ax1.xaxis.set_major_locator(MaxNLocator(integer=True)) # only show integer labels on x-axis + + # Create a second x-axis for tokens seen + ax2 = ax1.twiny() # Create a second x-axis that shares the same y-axis + ax2.plot(tokens_seen, train_losses, alpha=0) # Invisible plot for aligning ticks + ax2.set_xlabel("Tokens seen") + + fig.tight_layout() # Adjust layout to make room + print(f"Plot saved as {plot_name}") + plt.savefig(plot_name) + # plt.show() + + +def main(mask_instructions=False, alpaca52k=False, phi3_prompt=False, lora=False): + ####################################### + # Print package versions + ####################################### + print() + pkgs = [ + "matplotlib", # Plotting library + "tiktoken", # Tokenizer + "torch", # Deep learning library + "tqdm", # Progress bar + "tensorflow", # For OpenAI's pretrained weights + ] + for p in pkgs: + print(f"{p} version: {version(p)}") + print(50*"-") + + ####################################### + # Download and prepare dataset + ####################################### + file_path = "instruction-data.json" + + if alpaca52k: + url = "https://raw.githubusercontent.com/tatsu-lab/stanford_alpaca/main/alpaca_data.json" + else: + url = "https://raw.githubusercontent.com/rasbt/LLMs-from-scratch/main/ch07/01_main-chapter-code/instruction-data.json" + data = download_and_load_file(file_path, url) + + train_portion = int(len(data) * 0.85) # 85% for training + test_portion = int(len(data) * 0.1) # 10% for testing + + train_data = data[:train_portion] + test_data = data[train_portion:train_portion + test_portion] + val_data = data[train_portion + test_portion:] + + print("Training set length:", len(train_data)) + print("Validation set length:", len(val_data)) + print("Test set length:", len(test_data)) + print(50*"-") + + tokenizer = tiktoken.get_encoding("gpt2") + device = torch.device("cuda" if torch.cuda.is_available() else "cpu") + print("Device:", device) + print(50*"-") + + if alpaca52k: + allowed_max_length = 512 + else: + allowed_max_length = 1024 + + if mask_instructions and phi3_prompt: + raise ValueError("Simultaneous support for instruction masking and the Phi-3 prompt template has not been implemented, yet.") + + if mask_instructions: + customized_collate_fn = partial(custom_collate_with_masking_fn, device=device, allowed_max_length=allowed_max_length) + CustomDataset = InstructionDatasetWithMasking + elif phi3_prompt: + customized_collate_fn = partial(custom_collate_fn, device=device, allowed_max_length=allowed_max_length) + CustomDataset = InstructionDatasetPhi + else: + customized_collate_fn = partial(custom_collate_fn, device=device, allowed_max_length=allowed_max_length) + CustomDataset = InstructionDataset + + num_workers = 0 + + if alpaca52k: + batch_size = 4 + else: + batch_size = 8 + + torch.manual_seed(123) + + train_dataset = CustomDataset(train_data, tokenizer) + train_loader = DataLoader( + train_dataset, + batch_size=batch_size, + collate_fn=customized_collate_fn, + shuffle=True, + drop_last=True, + num_workers=num_workers + ) + + val_dataset = CustomDataset(val_data, tokenizer) + val_loader = DataLoader( + val_dataset, + batch_size=batch_size, + collate_fn=customized_collate_fn, + shuffle=False, + drop_last=False, + num_workers=num_workers + ) + + ####################################### + # Load pretrained model + ####################################### + BASE_CONFIG = { + "vocab_size": 50257, # Vocabulary size + "context_length": 1024, # Context length + "drop_rate": 0.0, # Dropout rate + "qkv_bias": True # Query-key-value bias + } + + model_configs = { + "gpt2-small (124M)": {"emb_dim": 768, "n_layers": 12, "n_heads": 12}, + "gpt2-medium (355M)": {"emb_dim": 1024, "n_layers": 24, "n_heads": 16}, + "gpt2-large (774M)": {"emb_dim": 1280, "n_layers": 36, "n_heads": 20}, + "gpt2-xl (1558M)": {"emb_dim": 1600, "n_layers": 48, "n_heads": 25}, + } + + CHOOSE_MODEL = "gpt2-medium (355M)" + + BASE_CONFIG.update(model_configs[CHOOSE_MODEL]) + + model_size = CHOOSE_MODEL.split(" ")[-1].lstrip("(").rstrip(")") + settings, params = download_and_load_gpt2(model_size=model_size, models_dir="gpt2") + + model = GPTModel(BASE_CONFIG) + load_weights_into_gpt(model, params) + model.eval() + model.to(device) + + print("Loaded model:", CHOOSE_MODEL) + print(50*"-") + + if lora: + total_params = sum(p.numel() for p in model.parameters() if p.requires_grad) + print(f"Total trainable parameters before: {total_params:,}") + + for param in model.parameters(): + param.requires_grad = False + + total_params = sum(p.numel() for p in model.parameters() if p.requires_grad) + print(f"Total trainable parameters after: {total_params:,}") + replace_linear_with_lora(model, rank=16, alpha=16) + + total_params = sum(p.numel() for p in model.parameters() if p.requires_grad) + print(f"Total trainable LoRA parameters: {total_params:,}") + model.to(device) + + ####################################### + # Finetuning the model + ####################################### + print("Initial losses") + with torch.no_grad(): + train_loss = calc_loss_loader(train_loader, model, device, num_batches=5) + val_loss = calc_loss_loader(val_loader, model, device, num_batches=5) + + print(" Training loss:", train_loss) + print(" Validation loss:", val_loss) + + start_time = time.time() + + num_epochs = 2 + optimizer = torch.optim.AdamW(model.parameters(), lr=0.00005, weight_decay=0.1) + + torch.manual_seed(123) + + start_context = format_input_phi(val_data[0]) if phi3_prompt else format_input(val_data[0]) + + train_losses, val_losses, tokens_seen = train_model_simple( + model, train_loader, val_loader, optimizer, device, + num_epochs=num_epochs, eval_freq=5, eval_iter=5, + start_context=start_context, tokenizer=tokenizer + ) + + end_time = time.time() + execution_time_minutes = (end_time - start_time) / 60 + print(f"Training completed in {execution_time_minutes:.2f} minutes.") + + epochs_tensor = torch.linspace(0, num_epochs, len(train_losses)) + + plot_name = "loss-plot.pdf" + if mask_instructions: + plot_name = plot_name.replace(".pdf", "-mask-instructions.pdf") + if alpaca52k: + plot_name = plot_name.replace(".pdf", "-alpaca52k.pdf") + if phi3_prompt: + plot_name = plot_name.replace(".pdf", "-phi3-prompt.pdf") + if lora: + plot_name = plot_name.replace(".pdf", "-lora.pdf") + if not any([mask_instructions, alpaca52k, phi3_prompt, lora]): + plot_name = plot_name.replace(".pdf", "-baseline.pdf") + + plot_losses(epochs_tensor, tokens_seen, train_losses, val_losses, plot_name) + print(50*"-") + + ####################################### + # Saving results + ####################################### + print("Generating responses") + for i, entry in tqdm(enumerate(test_data), total=len(test_data)): + + input_text = format_input_phi(entry) if phi3_prompt else format_input(entry) + + token_ids = generate( + model=model, + idx=text_to_token_ids(input_text, tokenizer).to(device), + max_new_tokens=256, + context_size=BASE_CONFIG["context_length"], + eos_id=50256 + ) + generated_text = token_ids_to_text(token_ids, tokenizer) + + if phi3_prompt: + response_text = generated_text[len(input_text):].replace("<|assistant|>:", "").strip() + else: + response_text = generated_text[len(input_text):].replace("### Response:", "").strip() + + test_data[i]["model_response"] = response_text + + test_data_path = "instruction-data-with-response.json" + file_name = f"{re.sub(r'[ ()]', '', CHOOSE_MODEL) }-sft.pth" + + if mask_instructions: + test_data_path = test_data_path.replace(".json", "-mask-instructions.json") + file_name = file_name.replace(".pth", "-mask-instructions.pth") + if alpaca52k: + test_data_path = test_data_path.replace(".json", "-alpaca52k.json") + file_name = file_name.replace(".pth", "-alpaca52k.pth") + if phi3_prompt: + test_data_path = test_data_path.replace(".json", "-phi3-prompt.json") + file_name = file_name.replace(".pth", "-phi3-prompt.pth") + if lora: + test_data_path = test_data_path.replace(".json", "-lora.json") + file_name = file_name.replace(".pth", "-lora.pth") + if not any([mask_instructions, alpaca52k, phi3_prompt, lora]): + test_data_path = test_data_path.replace(".json", "-baseline.json") + file_name = file_name.replace(".pth", "-baseline.pth") + + with open(test_data_path, "w") as file: + json.dump(test_data, file, indent=4) # "indent" for pretty-printing + print(f"Responses saved as {test_data_path}") + + torch.save(model.state_dict(), file_name) + print(f"Model saved as {file_name}") + + +if __name__ == "__main__": + + import argparse + + parser = argparse.ArgumentParser( + description="Instruction finetune a GPT model" + ) + options = {"baseline", "mask_instructions", "alpaca_52k", "phi3_prompt", "lora"} + parser.add_argument( + "--exercise_solution", + type=str, + default="last_block", + help=( + f"Which experiment to run. Options: {options}." + ) + ) + args = parser.parse_args() + + if args.exercise_solution == "baseline": + main() + elif args.exercise_solution == "mask_instructions": + main(mask_instructions=True) + elif args.exercise_solution == "alpaca_52k": + main(alpaca52k=True) + elif args.exercise_solution == "phi3_prompt": + main(phi3_prompt=True) + elif args.exercise_solution == "lora": + main(lora=True) + else: + raise ValueError(f"{args.exercise_solution} is not a valid --args.exercise_solution option. Options: {options}") diff --git a/Codes/ch07/01_main-chapter-code/gpt_download.py b/Codes/ch07/01_main-chapter-code/gpt_download.py new file mode 100644 index 0000000..aa0ea1e --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/gpt_download.py @@ -0,0 +1,142 @@ +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch + + +import os +import urllib.request + +# import requests +import json +import numpy as np +import tensorflow as tf +from tqdm import tqdm + + +def download_and_load_gpt2(model_size, models_dir): + # Validate model size + allowed_sizes = ("124M", "355M", "774M", "1558M") + if model_size not in allowed_sizes: + raise ValueError(f"Model size not in {allowed_sizes}") + + # Define paths + model_dir = os.path.join(models_dir, model_size) + base_url = "https://openaipublic.blob.core.windows.net/gpt-2/models" + filenames = [ + "checkpoint", "encoder.json", "hparams.json", + "model.ckpt.data-00000-of-00001", "model.ckpt.index", + "model.ckpt.meta", "vocab.bpe" + ] + + # Download files + os.makedirs(model_dir, exist_ok=True) + for filename in filenames: + file_url = os.path.join(base_url, model_size, filename) + file_path = os.path.join(model_dir, filename) + download_file(file_url, file_path) + + # Load settings and params + tf_ckpt_path = tf.train.latest_checkpoint(model_dir) + settings = json.load(open(os.path.join(model_dir, "hparams.json"))) + params = load_gpt2_params_from_tf_ckpt(tf_ckpt_path, settings) + + return settings, params + + +def download_file(url, destination): + # Send a GET request to download the file + + try: + with urllib.request.urlopen(url) as response: + # Get the total file size from headers, defaulting to 0 if not present + file_size = int(response.headers.get("Content-Length", 0)) + + # Check if file exists and has the same size + if os.path.exists(destination): + file_size_local = os.path.getsize(destination) + if file_size == file_size_local: + print(f"File already exists and is up-to-date: {destination}") + return + + # Define the block size for reading the file + block_size = 1024 # 1 Kilobyte + + # Initialize the progress bar with total file size + progress_bar_description = os.path.basename(url) # Extract filename from URL + with tqdm(total=file_size, unit="iB", unit_scale=True, desc=progress_bar_description) as progress_bar: + # Open the destination file in binary write mode + with open(destination, "wb") as file: + # Read the file in chunks and write to destination + while True: + chunk = response.read(block_size) + if not chunk: + break + file.write(chunk) + progress_bar.update(len(chunk)) # Update progress bar + except urllib.error.HTTPError: + s = ( + f"The specified URL ({url}) is incorrect, the internet connection cannot be established," + "\nor the requested file is temporarily unavailable.\nPlease visit the following website" + " for help: https://github.com/rasbt/LLMs-from-scratch/discussions/273") + print(s) + + +# Alternative way using `requests` +""" +def download_file(url, destination): + # Send a GET request to download the file in streaming mode + response = requests.get(url, stream=True) + + # Get the total file size from headers, defaulting to 0 if not present + file_size = int(response.headers.get("content-length", 0)) + + # Check if file exists and has the same size + if os.path.exists(destination): + file_size_local = os.path.getsize(destination) + if file_size == file_size_local: + print(f"File already exists and is up-to-date: {destination}") + return + + # Define the block size for reading the file + block_size = 1024 # 1 Kilobyte + + # Initialize the progress bar with total file size + progress_bar_description = url.split("/")[-1] # Extract filename from URL + with tqdm(total=file_size, unit="iB", unit_scale=True, desc=progress_bar_description) as progress_bar: + # Open the destination file in binary write mode + with open(destination, "wb") as file: + # Iterate over the file data in chunks + for chunk in response.iter_content(block_size): + progress_bar.update(len(chunk)) # Update progress bar + file.write(chunk) # Write the chunk to the file +""" + + +def load_gpt2_params_from_tf_ckpt(ckpt_path, settings): + # Initialize parameters dictionary with empty blocks for each layer + params = {"blocks": [{} for _ in range(settings["n_layer"])]} + + # Iterate over each variable in the checkpoint + for name, _ in tf.train.list_variables(ckpt_path): + # Load the variable and remove singleton dimensions + variable_array = np.squeeze(tf.train.load_variable(ckpt_path, name)) + + # Process the variable name to extract relevant parts + variable_name_parts = name.split("/")[1:] # Skip the 'model/' prefix + + # Identify the target dictionary for the variable + target_dict = params + if variable_name_parts[0].startswith("h"): + layer_number = int(variable_name_parts[0][1:]) + target_dict = params["blocks"][layer_number] + + # Recursively access or create nested dictionaries + for key in variable_name_parts[1:-1]: + target_dict = target_dict.setdefault(key, {}) + + # Assign the variable array to the last key + last_key = variable_name_parts[-1] + target_dict[last_key] = variable_array + + return params diff --git a/Codes/ch07/01_main-chapter-code/gpt_instruction_finetuning.py b/Codes/ch07/01_main-chapter-code/gpt_instruction_finetuning.py new file mode 100644 index 0000000..6bc6429 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/gpt_instruction_finetuning.py @@ -0,0 +1,351 @@ +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch +# +# A minimal instruction finetuning file based on the code in chapter 7 + +from functools import partial +from importlib.metadata import version +import json +import os +import re +import time +import urllib + +import matplotlib.pyplot as plt +import tiktoken +import torch +from torch.utils.data import Dataset, DataLoader +from tqdm import tqdm + +# Import from local files in this folder +from gpt_download import download_and_load_gpt2 +from previous_chapters import ( + calc_loss_loader, + generate, + GPTModel, + load_weights_into_gpt, + text_to_token_ids, + train_model_simple, + token_ids_to_text +) + + +class InstructionDataset(Dataset): + def __init__(self, data, tokenizer): + self.data = data + + # Pre-tokenize texts + self.encoded_texts = [] + for entry in data: + instruction_plus_input = format_input(entry) + response_text = f"\n\n### Response:\n{entry['output']}" + full_text = instruction_plus_input + response_text + self.encoded_texts.append( + tokenizer.encode(full_text) + ) + + def __getitem__(self, index): + return self.encoded_texts[index] + + def __len__(self): + return len(self.data) + + +def custom_collate_fn( + batch, + pad_token_id=50256, + ignore_index=-100, + allowed_max_length=None, + device="cpu" +): + # Find the longest sequence in the batch + batch_max_length = max(len(item)+1 for item in batch) + + # Pad and prepare inputs and targets + inputs_lst, targets_lst = [], [] + + for item in batch: + new_item = item.copy() + # Add an <|endoftext|> token + new_item += [pad_token_id] + # Pad sequences to max_length + padded = new_item + [pad_token_id] * (batch_max_length - len(new_item)) + inputs = torch.tensor(padded[:-1]) # Truncate the last token for inputs + targets = torch.tensor(padded[1:]) # Shift +1 to the right for targets + + # New: Replace all but the first padding tokens in targets by ignore_index + mask = targets == pad_token_id + indices = torch.nonzero(mask).squeeze() + if indices.numel() > 1: + targets[indices[1:]] = ignore_index + + # New: Optionally truncate to maximum sequence length + if allowed_max_length is not None: + inputs = inputs[:allowed_max_length] + targets = targets[:allowed_max_length] + + inputs_lst.append(inputs) + targets_lst.append(targets) + + # Convert list of inputs and targets to tensors and transfer to target device + inputs_tensor = torch.stack(inputs_lst).to(device) + targets_tensor = torch.stack(targets_lst).to(device) + + return inputs_tensor, targets_tensor + + +def download_and_load_file(file_path, url): + + if not os.path.exists(file_path): + with urllib.request.urlopen(url) as response: + text_data = response.read().decode("utf-8") + with open(file_path, "w", encoding="utf-8") as file: + file.write(text_data) + else: + with open(file_path, "r", encoding="utf-8") as file: + text_data = file.read() + + with open(file_path, "r") as file: + data = json.load(file) + + return data + + +def format_input(entry): + instruction_text = ( + f"Below is an instruction that describes a task. " + f"Write a response that appropriately completes the request." + f"\n\n### Instruction:\n{entry['instruction']}" + ) + + input_text = f"\n\n### Input:\n{entry['input']}" if entry["input"] else "" + + return instruction_text + input_text + + +def plot_losses(epochs_seen, tokens_seen, train_losses, val_losses): + fig, ax1 = plt.subplots(figsize=(12, 6)) + + # Plot training and validation loss against epochs + ax1.plot(epochs_seen, train_losses, label="Training loss") + ax1.plot(epochs_seen, val_losses, linestyle="-.", label="Validation loss") + ax1.set_xlabel("Epochs") + ax1.set_ylabel("Loss") + ax1.legend(loc="upper right") + + # Create a second x-axis for tokens seen + ax2 = ax1.twiny() # Create a second x-axis that shares the same y-axis + ax2.plot(tokens_seen, train_losses, alpha=0) # Invisible plot for aligning ticks + ax2.set_xlabel("Tokens seen") + + fig.tight_layout() # Adjust layout to make room + plot_name = "loss-plot-standalone.pdf" + print(f"Plot saved as {plot_name}") + plt.savefig(plot_name) + # plt.show() + + +def main(test_mode=False): + ####################################### + # Print package versions + ####################################### + print() + pkgs = [ + "matplotlib", # Plotting library + "tiktoken", # Tokenizer + "torch", # Deep learning library + "tqdm", # Progress bar + "tensorflow", # For OpenAI's pretrained weights + ] + for p in pkgs: + print(f"{p} version: {version(p)}") + print(50*"-") + + ####################################### + # Download and prepare dataset + ####################################### + file_path = "instruction-data.json" + url = "https://raw.githubusercontent.com/rasbt/LLMs-from-scratch/main/ch07/01_main-chapter-code/instruction-data.json" + data = download_and_load_file(file_path, url) + + train_portion = int(len(data) * 0.85) # 85% for training + test_portion = int(len(data) * 0.1) # 10% for testing + + train_data = data[:train_portion] + test_data = data[train_portion:train_portion + test_portion] + val_data = data[train_portion + test_portion:] + + # Use very small subset for testing purposes + if args.test_mode: + train_data = train_data[:10] + val_data = val_data[:10] + test_data = test_data[:10] + + print("Training set length:", len(train_data)) + print("Validation set length:", len(val_data)) + print("Test set length:", len(test_data)) + print(50*"-") + + tokenizer = tiktoken.get_encoding("gpt2") + device = torch.device("cuda" if torch.cuda.is_available() else "cpu") + print("Device:", device) + print(50*"-") + + customized_collate_fn = partial(custom_collate_fn, device=device, allowed_max_length=1024) + + num_workers = 0 + batch_size = 8 + + torch.manual_seed(123) + + train_dataset = InstructionDataset(train_data, tokenizer) + train_loader = DataLoader( + train_dataset, + batch_size=batch_size, + collate_fn=customized_collate_fn, + shuffle=True, + drop_last=True, + num_workers=num_workers + ) + + val_dataset = InstructionDataset(val_data, tokenizer) + val_loader = DataLoader( + val_dataset, + batch_size=batch_size, + collate_fn=customized_collate_fn, + shuffle=False, + drop_last=False, + num_workers=num_workers + ) + + ####################################### + # Load pretrained model + ####################################### + + # Small GPT model for testing purposes + if args.test_mode: + BASE_CONFIG = { + "vocab_size": 50257, + "context_length": 120, + "drop_rate": 0.0, + "qkv_bias": False, + "emb_dim": 12, + "n_layers": 1, + "n_heads": 2 + } + model = GPTModel(BASE_CONFIG) + model.eval() + device = "cpu" + CHOOSE_MODEL = "Small test model" + + # Code as it is used in the main chapter + else: + BASE_CONFIG = { + "vocab_size": 50257, # Vocabulary size + "context_length": 1024, # Context length + "drop_rate": 0.0, # Dropout rate + "qkv_bias": True # Query-key-value bias + } + + model_configs = { + "gpt2-small (124M)": {"emb_dim": 768, "n_layers": 12, "n_heads": 12}, + "gpt2-medium (355M)": {"emb_dim": 1024, "n_layers": 24, "n_heads": 16}, + "gpt2-large (774M)": {"emb_dim": 1280, "n_layers": 36, "n_heads": 20}, + "gpt2-xl (1558M)": {"emb_dim": 1600, "n_layers": 48, "n_heads": 25}, + } + + CHOOSE_MODEL = "gpt2-medium (355M)" + + BASE_CONFIG.update(model_configs[CHOOSE_MODEL]) + + model_size = CHOOSE_MODEL.split(" ")[-1].lstrip("(").rstrip(")") + settings, params = download_and_load_gpt2(model_size=model_size, models_dir="gpt2") + + model = GPTModel(BASE_CONFIG) + load_weights_into_gpt(model, params) + model.eval() + model.to(device) + + print("Loaded model:", CHOOSE_MODEL) + print(50*"-") + + ####################################### + # Finetuning the model + ####################################### + print("Initial losses") + with torch.no_grad(): + train_loss = calc_loss_loader(train_loader, model, device, num_batches=5) + val_loss = calc_loss_loader(val_loader, model, device, num_batches=5) + + print(" Training loss:", train_loss) + print(" Validation loss:", val_loss) + + start_time = time.time() + optimizer = torch.optim.AdamW(model.parameters(), lr=0.00005, weight_decay=0.1) + + num_epochs = 2 + + torch.manual_seed(123) + train_losses, val_losses, tokens_seen = train_model_simple( + model, train_loader, val_loader, optimizer, device, + num_epochs=num_epochs, eval_freq=5, eval_iter=5, + start_context=format_input(val_data[0]), tokenizer=tokenizer + ) + + end_time = time.time() + execution_time_minutes = (end_time - start_time) / 60 + print(f"Training completed in {execution_time_minutes:.2f} minutes.") + + epochs_tensor = torch.linspace(0, num_epochs, len(train_losses)) + plot_losses(epochs_tensor, tokens_seen, train_losses, val_losses) + print(50*"-") + + ####################################### + # Saving results + ####################################### + print("Generating responses") + for i, entry in tqdm(enumerate(test_data), total=len(test_data)): + + input_text = format_input(entry) + + token_ids = generate( + model=model, + idx=text_to_token_ids(input_text, tokenizer).to(device), + max_new_tokens=256, + context_size=BASE_CONFIG["context_length"], + eos_id=50256 + ) + generated_text = token_ids_to_text(token_ids, tokenizer) + response_text = generated_text[len(input_text):].replace("### Response:", "").strip() + + test_data[i]["model_response"] = response_text + + test_data_path = "instruction-data-with-response-standalone.json" + with open(test_data_path, "w") as file: + json.dump(test_data, file, indent=4) # "indent" for pretty-printing + print(f"Responses saved as {test_data_path}") + + file_name = f"{re.sub(r'[ ()]', '', CHOOSE_MODEL) }-sft-standalone.pth" + torch.save(model.state_dict(), file_name) + print(f"Model saved as {file_name}") + + +if __name__ == "__main__": + + import argparse + + parser = argparse.ArgumentParser( + description="Finetune a GPT model for classification" + ) + parser.add_argument( + "--test_mode", + default=False, + action="store_true", + help=("This flag runs the model in test mode for internal testing purposes. " + "Otherwise, it runs the model as it is used in the chapter (recommended).") + ) + args = parser.parse_args() + + main(args.test_mode) diff --git a/Codes/ch07/01_main-chapter-code/instruction-data-with-response.json b/Codes/ch07/01_main-chapter-code/instruction-data-with-response.json new file mode 100644 index 0000000..1d19190 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/instruction-data-with-response.json @@ -0,0 +1,662 @@ +[ + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The car is very fast.", + "output": "The car is as fast as lightning.", + "model_response": "The car is as fast as a bullet." + }, + { + "instruction": "What type of cloud is typically associated with thunderstorms?", + "input": "", + "output": "The type of cloud typically associated with thunderstorms is cumulonimbus.", + "model_response": "The type of cloud associated with thunderstorms is a cumulus cloud." + }, + { + "instruction": "Name the author of 'Pride and Prejudice'.", + "input": "", + "output": "Jane Austen.", + "model_response": "The author of 'Pride and Prejudice' is Jane Austen." + }, + { + "instruction": "What is the periodic symbol for chlorine?", + "input": "", + "output": "The periodic symbol for chlorine is Cl.", + "model_response": "The periodic symbol for chlorine is C." + }, + { + "instruction": "Correct the punctuation in the sentence.", + "input": "Its time to go home.", + "output": "The corrected sentence should be: 'It's time to go home.'", + "model_response": "It's time to go home." + }, + { + "instruction": "Rewrite the sentence.", + "input": "The lecture was delivered in a clear manner.", + "output": "The lecture was delivered clearly.", + "model_response": "The lecture was delivered in a clear manner." + }, + { + "instruction": "Generate a humorous anecdote.", + "input": "", + "output": "Why was the math book sad? Because it had too many problems!", + "model_response": "The cat jumped over the fence." + }, + { + "instruction": "Identify the correct spelling: 'recieve' or 'receive'.", + "input": "", + "output": "The correct spelling is 'receive'.", + "model_response": "The correct spelling is 'receive'." + }, + { + "instruction": "Create a sentence using the word 'nostalgia'.", + "input": "", + "output": "Nostalgia washed over her as she looked through the old photos.", + "model_response": "Nostalgia is a strong emotion that can be felt in the past." + }, + { + "instruction": "Classify the following numbers as prime or composite.", + "input": ": 11, 14, 19.", + "output": "Prime numbers: 11, 19\nComposite numbers: 14", + "model_response": "Prime numbers: 11, 14, 19.Composite numbers: 14, 19." + }, + { + "instruction": "What is the capital of Denmark?", + "input": "", + "output": "The capital of Denmark is Copenhagen.", + "model_response": "The capital of Denmark is Copenhagen." + }, + { + "instruction": "What is the opposite of 'wet'?", + "input": "", + "output": "The opposite of 'wet' is 'dry'.", + "model_response": "The opposite of 'wet' is 'dry'." + }, + { + "instruction": "Identify the type of sentence.", + "input": "Did you finish the report?", + "output": "The type of sentence is interrogative.", + "model_response": "The type of sentence is imperative." + }, + { + "instruction": "What is the chemical symbol for mercury?", + "input": "", + "output": "Hg.", + "model_response": "The chemical symbol for mercury is Hg." + }, + { + "instruction": "Describe the process of active transport in cells.", + "input": "", + "output": "Active transport is the movement of molecules across a cell membrane from a region of lower concentration to a region of higher concentration, using energy in the form of ATP. This process is crucial for maintaining cellular homeostasis.", + "model_response": "Transport is the process by which a substance moves from one place to another by chemical change. It typically occurs by chemical reactions between atoms or molecules." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Mercury, oxygen, wood", + "output": "Mercury - Liquid\nOxygen - Gas\nWood - Solid", + "model_response": ", water\n\n\nSolid: Carbon\nLiquefied Carbon: Oxygen\nGas: Carbon dioxide\nGaseous Carbon: Carbon monoxide" + }, + { + "instruction": "Convert 3 kilometers to meters.", + "input": "", + "output": "3 kilometers is 3000 meters.", + "model_response": "3 kilometers is 3000 meters." + }, + { + "instruction": "Convert the sentence to use an indefinite pronoun.", + "input": "Someone left a note.", + "output": "A note was left by someone.", + "model_response": "A note was left by someone." + }, + { + "instruction": "Provide a synonym for 'excited'.", + "input": "", + "output": "A synonym for 'excited' is 'thrilled'.", + "model_response": "A synonym for 'excited' is 'enthusiastic'." + }, + { + "instruction": "Generate a sentence that follows the pattern: \"Never have I ever _____ without _____\"", + "input": "", + "output": "Never have I ever traveled without a map.", + "model_response": "Never have I ever had without having." + }, + { + "instruction": "Pick out the adjective from the following list.", + "input": "run, tall, quickly", + "output": "The correct adjective from the list is 'tall.'", + "model_response": "The adjective from the list is 'quick'." + }, + { + "instruction": "Convert 1000 grams to kilograms.", + "input": "", + "output": "1000 grams is equal to 1 kilogram.", + "model_response": "1000 grams is 0.1 kilograms." + }, + { + "instruction": "What is the opposite of 'deep'?", + "input": "", + "output": "The opposite of 'deep' is 'shallow'.", + "model_response": "The opposite of 'deep' is 'light'." + }, + { + "instruction": "Categorize the given list of animals.", + "input": "Shark, Dolphin, Trout", + "output": "Fish: Shark, Trout\nMammals: Dolphin", + "model_response": "Animals: Shark, Dolphin, Trout" + }, + { + "instruction": "Translate 'library' into Spanish.", + "input": "", + "output": "The Spanish word for 'library' is 'biblioteca'.", + "model_response": "The Spanish translation of 'library' is 'l\u00edmite'." + }, + { + "instruction": "Define the term 'hyperbole'.", + "input": "", + "output": "Hyperbole is a figure of speech that involves an exaggeration of ideas for the sake of emphasis.", + "model_response": "Hyperbole is a word used to describe a statement that is exaggerated or exaggerated in meaning." + }, + { + "instruction": "Translate 'Hello' into Russian.", + "input": "", + "output": "\u041f\u0440\u0438\u0432\u0435\u0442 (Privet)", + "model_response": "The Russian translation of 'Hello' is '\u0420\u0443\u0441\u0441\u043a\u0438\u0439 \u043f\u0440\u043e\u0438\u0437\u0432\u0435\u0434\u0435\u043d\u0438\u0439' (Vyacheslav Krasnoyarski)." + }, + { + "instruction": "Define the term 'kinetic energy'.", + "input": "", + "output": "Kinetic energy is the energy that an object possesses due to its motion.", + "model_response": "Kinetic energy is the force that causes an object to accelerate or decelerate in response to an applied force." + }, + { + "instruction": "What is the opposite of 'hot'?", + "input": "", + "output": "The opposite of 'hot' is 'cold'.", + "model_response": "The opposite of 'hot' is 'cold'." + }, + { + "instruction": "Convert 5 miles to kilometers.", + "input": "", + "output": "5 miles is approximately 8.05 kilometers.", + "model_response": "5 miles is 5000 meters." + }, + { + "instruction": "What is the chemical formula for magnesium sulfate?", + "input": "", + "output": "The chemical formula for magnesium sulfate is MgSO4.", + "model_response": "The chemical formula for magnesium sulfate is MgSO4." + }, + { + "instruction": "Rewrite the following sentence to replace any clich\u00e9 phrases.", + "input": "It's a piece of cake", + "output": "It's very easy.", + "model_response": ".\n\n\nA piece of cake is a cake." + }, + { + "instruction": "List five different types of vegetables.", + "input": "", + "output": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato\n5. Spinach", + "model_response": "1. Cucumber\n2. Broccoli\n3. Spinach\n4. Tomato\n5. Cucumber" + }, + { + "instruction": "Convert 7 kilometers to meters.", + "input": "", + "output": "7 kilometers is 7000 meters.", + "model_response": "7 kilometers is 7.048 meters." + }, + { + "instruction": "What is the opposite of 'heavy'?", + "input": "", + "output": "The opposite of 'heavy' is 'light'.", + "model_response": "The opposite of 'heavy' is 'light'." + }, + { + "instruction": "What is the past tense of 'sing'?", + "input": "", + "output": "The past tense of 'sing' is 'sang.'", + "model_response": "The past tense of 'sing' is 'to sing.'" + }, + { + "instruction": "What is the molecular formula for carbon dioxide?", + "input": "", + "output": "The molecular formula for carbon dioxide is CO2.", + "model_response": "The molecular formula for carbon dioxide is CO2." + }, + { + "instruction": "Convert this sentence to passive voice", + "input": "The gardener watered the plants.", + "output": "The plants were watered by the gardener.", + "model_response": "The plants were watered by the gardener." + }, + { + "instruction": "What is the past tense of 'throw'?", + "input": "", + "output": "The past tense of 'throw' is 'threw'.", + "model_response": "The past tense of 'throw' is 'to throw'." + }, + { + "instruction": "Explain what a sonnet is.", + "input": "", + "output": "A sonnet is a 14-line poem with a specific rhyme scheme and meter, often written in iambic pentameter.", + "model_response": "A sonnet is a poem composed in the first person." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "The company is known for its innovative products.", + "model_response": "She was very innovative and always had a new idea." + }, + { + "instruction": "Provide the plural form of 'cactus'.", + "input": "", + "output": "The plural form of 'cactus' is 'cacti'.", + "model_response": "The plural form of 'cactus' is 'cacti'." + }, + { + "instruction": "Translate the phrase 'Where is the bathroom?' into German.", + "input": "", + "output": "The German translation of 'Where is the bathroom?' is 'Wo ist die Toilette?'", + "model_response": "The German translation of 'Where is the bathroom?' is 'Wie es \u00fcber das B\u00fccher?'." + }, + { + "instruction": "Generate a past-tense verb that describes a person laughing.", + "input": "", + "output": "Laughed.", + "model_response": "She laughed." + }, + { + "instruction": "Generate a sentence using the word 'transient.'", + "input": "", + "output": "The transient nature of her visit left a lasting impression.", + "model_response": "She was seen by a passing car as it passed by." + }, + { + "instruction": "Generate a sentence using the word 'optimistic'.", + "input": "", + "output": "He remained optimistic despite the challenges he faced.", + "model_response": "She was optimistic about the future." + }, + { + "instruction": "Re-word this sentence using an indirect question.", + "input": "What time is the meeting?", + "output": "Could you tell me what time the meeting is?", + "model_response": "The meeting is at 7:00 PM." + }, + { + "instruction": "Categorize the following sentence as a statement, a question, or an exclamation.", + "input": "What a beautiful day!", + "output": "Exclamation.", + "model_response": "A statement could be classified as a question." + }, + { + "instruction": "What is the opposite of 'rich'?", + "input": "", + "output": "The opposite of 'rich' is 'poor'.", + "model_response": "The opposite of 'rich' is 'poor'." + }, + { + "instruction": "Find a synonym for the given verb.", + "input": "Begin", + "output": "Commence", + "model_response": "Begin\n\n### Input:\nBegin\n\n\nBegin\n\n### Input:\nBegin\n\nBegin\n### Input:\nBegin\n\nBegin" + }, + { + "instruction": "Edit the given text to ensure all plural nouns are spelled correctly.", + "input": "The birds sings beautiful songs.", + "output": "The birds sing beautiful songs.", + "model_response": "The birds sings beautiful songs." + }, + { + "instruction": "Transform the following sentence into a question using \"could.\"", + "input": "You can help me tomorrow.", + "output": "Could you help me tomorrow?", + "model_response": "Could you help me tomorrow?" + }, + { + "instruction": "Classify the following items: bicycle, rose, tiger.", + "input": "", + "output": "Vehicles: Bicycle\nPlants: Rose\nAnimals: Tiger", + "model_response": "Animals: Tiger\nBicyclist: Tiger\nRose: Rose\nBike: Tiger" + }, + { + "instruction": "Define the term 'irony'.", + "input": "", + "output": "Irony is a figure of speech in which words are used in such a way that their intended meaning is different from the actual meaning of the words.", + "model_response": "An irony is a statement that is at odds with the truth." + }, + { + "instruction": "Translate 'Welcome' into German.", + "input": "", + "output": "The German translation of 'Welcome' is 'Willkommen'.", + "model_response": "The German translation of 'Welcome' is 'Wie es tut mir leid'." + }, + { + "instruction": "Explain the primary function of the human heart.", + "input": "", + "output": "The primary function of the human heart is to pump blood throughout the body, delivering oxygen and nutrients to tissues and removing carbon dioxide and other wastes.", + "model_response": "The primary function of the human heart is to pump blood to the brain and to supply the muscles and nerves that control movement." + }, + { + "instruction": "Reword the following sentence to the future tense.", + "input": "He is reading a novel inspired by his grandmother.", + "output": "He will be reading a novel inspired by his grandmother.", + "model_response": "He is reading a novel inspired by his grandmother." + }, + { + "instruction": "Convert the given sentence into active voice.", + "input": "The law was passed by the government.", + "output": "The government passed the law.", + "model_response": "The law was passed by the government." + }, + { + "instruction": "Create a sentence using the word 'inevitable'.", + "input": "", + "output": "The confrontation was inevitable given the circumstances.", + "model_response": "The storm was inevitable." + }, + { + "instruction": "Categorize the following sentence as either factual or opinion-based.", + "input": "Chocolate is the best dessert.", + "output": "Opinion-based.", + "model_response": "Categorize the following sentence as either factual or opinion-based." + }, + { + "instruction": "What is an antonym of 'old'?", + "input": "", + "output": "young.", + "model_response": "An antonym of 'old' is 'young'." + }, + { + "instruction": "Provide a synonym for 'hardworking'.", + "input": "", + "output": "A synonym for 'hardworking' is 'diligent'.", + "model_response": "A synonym for 'hardworking' is 'smart'." + }, + { + "instruction": "What is the boiling point of sulfur in Celsius?", + "input": "", + "output": "The boiling point of sulfur is 444.6 degrees Celsius.", + "model_response": "The boiling point of sulfur is -114.5 degrees Celsius." + }, + { + "instruction": "What is the plural form of 'child'?", + "input": "", + "output": "The plural form of 'child' is 'children'.", + "model_response": "The plural form of 'child' is 'chunk'." + }, + { + "instruction": "What is an antonym of 'complicated'?", + "input": "", + "output": "An antonym of 'complicated' is 'simple'.", + "model_response": "An antonym of 'complicated' is 'simple'." + }, + { + "instruction": "Name three forms of water.", + "input": "", + "output": "The three forms of water are solid (ice), liquid (water), and gas (steam).", + "model_response": "Three forms of water are rain, snow, and water." + }, + { + "instruction": "Rewrite this sentence as a question.", + "input": "The dog chased the cat.", + "output": "Did the dog chase the cat?", + "model_response": "What is the cat chasing?" + }, + { + "instruction": "Split the following sentence into two declarative sentences: 'The movie was long but interesting.'", + "input": "", + "output": "The movie was long. It was interesting.", + "model_response": "and 'The movie was interesting.'\n\n\nThe movie was long, but it was interesting." + }, + { + "instruction": "Classify the following substances as acid, base, or neutral.", + "input": "Lemon juice, Soap, Water", + "output": "Acid: Lemon juice\nBase: Soap\nNeutral: Water", + "model_response": "Acid: Soap\nBase: Soap\nNeutral: Water" + }, + { + "instruction": "What is a synonym of 'sad'?", + "input": "", + "output": "A synonym for 'sad' is 'unhappy'.", + "model_response": "A synonym for 'sad' is 'angry'." + }, + { + "instruction": "Correct any spelling mistakes in the given sentence.", + "input": "I prefer homemade cookies to store boaght.", + "output": "I prefer homemade cookies to store bought.", + "model_response": "I prefer homemade cookies to store boaght." + }, + { + "instruction": "Generate a sentence using the word 'transient'.", + "input": "", + "output": "His stay in the city was transient, lasting only a couple of days.", + "model_response": "She was seen by a passerby as she passed by." + }, + { + "instruction": "Translate 'I am lost' into Italian.", + "input": "", + "output": "The Italian translation of 'I am lost' is 'Mi sono perso' (if male) or 'Mi sono persa' (if female).", + "model_response": "Il pi\u00f9 am\u00e9ricaino." + }, + { + "instruction": "Classify this text as a technical document or a narrative.", + "input": "This manual provides instructions for installing the software.", + "output": "Technical document", + "model_response": "Technical document" + }, + { + "instruction": "Sort the following list in descending order.", + "input": "10, 2, 25, 16, 7", + "output": "25, 16, 10, 7, 2.", + "model_response": ", 3, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25" + }, + { + "instruction": "Translate 'Can I have some water?' into French.", + "input": "", + "output": "Puis-je avoir de l'eau?", + "model_response": "Can I have some water?" + }, + { + "instruction": "Create a simile with the word 'as cold as'.", + "input": "", + "output": "Her hands were as cold as ice.", + "model_response": "The temperature of 'as cold as' is -38 degrees Celsius." + }, + { + "instruction": "Classify the following words by their grammatical categories: swim, beautiful, quickly", + "input": "", + "output": "Swim: Verb\nBeautiful: Adjective\nQuickly: Adverb", + "model_response": "Classifying the following words by their grammatical categories: swim, beautiful, quickly" + }, + { + "instruction": "Calculate the density of an object with a mass of 15 grams and a volume of 5 cubic centimeters.", + "input": "", + "output": "The density of the object is 3 grams per cubic centimeter.", + "model_response": "The density of the object is 15 grams/cm\u00b3." + }, + { + "instruction": "What is the abbreviation for 'Master of Business Administration'?", + "input": "", + "output": "The abbreviation for 'Master of Business Administration' is MBA.", + "model_response": "The abbreviation for 'Master of Business Administration' is MBA." + }, + { + "instruction": "Convert the following number from Roman numerals: IX.", + "input": "", + "output": "The number IX in Roman numerals is 9.", + "model_response": "5 to decimal numbers: 0.5.\n\n\n0.5 is 0.5." + }, + { + "instruction": "What is the opposite of 'horizontal'?", + "input": "", + "output": "The opposite of 'horizontal' is 'vertical'.", + "model_response": "The opposite of 'horizontal' is 'vertical'." + }, + { + "instruction": "Translate 'Where can I buy tickets?' into Italian.", + "input": "", + "output": "The Italian translation for 'Where can I buy tickets?' is 'Dove posso comprare i biglietti?'", + "model_response": "The Italian translation of 'Where can I buy tickets?' is 'Ti amo'." + }, + { + "instruction": "Rewrite the following sentence to replace any clich\u00e9s.", + "input": "He was as cool as a cucumber.", + "output": "He remained very calm.", + "model_response": "He was as cool as a cucumber." + }, + { + "instruction": "Identify the main verb in the sentence.", + "input": "The dog barked loudly.", + "output": "The main verb in the sentence is 'barked'.", + "model_response": "The main verb in the sentence is 'bark'." + }, + { + "instruction": "Generate a sentence using the word 'elucidate'.", + "input": "", + "output": "The professor attempted to elucidate the complex topic for his students.", + "model_response": "The teacher had studied the book for a long time." + }, + { + "instruction": "Correct the sentence.", + "input": "Me and my friend went to the store.", + "output": "My friend and I went to the store.", + "model_response": "We went to the store." + }, + { + "instruction": "What is the formula for calculating work done?", + "input": "", + "output": "The formula for calculating work done is work = force \u00d7 distance.", + "model_response": "The formula for calculating work done is W=M*G*A." + }, + { + "instruction": "What is the chemical formula for ammonium nitrate?", + "input": "", + "output": "The chemical formula for ammonium nitrate is NH4NO3.", + "model_response": "The chemical formula for ammonium nitrate is NH3." + }, + { + "instruction": "What is the molecular formula for water?", + "input": "", + "output": "The molecular formula for water is H2O.", + "model_response": "The molecular formula for water is H2O2." + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The food was not good.", + "output": "The food could use some improvement.", + "model_response": "The food was not good." + }, + { + "instruction": "What is the opposite of 'lazy'?", + "input": "", + "output": "The opposite of 'lazy' is 'diligent'.", + "model_response": "The opposite of 'lazy' is 'attentive'." + }, + { + "instruction": "Name three essential vitamins for human health.", + "input": "", + "output": "1. Vitamin A\n2. Vitamin C\n3. Vitamin D", + "model_response": "1. B6\n2. B12\n3. B12\n### Instruction:\nWhat is the chemical symbol for iron?\n\n\nThe chemical symbol for iron is Fe." + }, + { + "instruction": "Define the term 'simile'.", + "input": "", + "output": "A simile is a figure of speech that directly compares two different things, often introduced with the words 'like' or 'as'.", + "model_response": "A simile is a figurative device used to describe a person or thing that is simile or similitude in form or function." + }, + { + "instruction": "What is the boiling point of chlorine in Celsius?", + "input": "", + "output": "The boiling point of chlorine is -34 degrees Celsius.", + "model_response": "The boiling point of chlorine is -196 degrees Celsius." + }, + { + "instruction": "Translate 'My name is' into French.", + "input": "", + "output": "The French translation of 'My name is' is 'Je m'appelle'.", + "model_response": "The French translation of 'My name is' is 'Comment \u00e7a va?'." + }, + { + "instruction": "Convert 200 centimeters to meters.", + "input": "", + "output": "200 centimeters is 2 meters.", + "model_response": "200 centimeters is 0.2 meters." + }, + { + "instruction": "What is the chemical symbol for zinc?", + "input": "", + "output": "Zn.", + "model_response": "The chemical symbol for zinc is Zn." + }, + { + "instruction": "What is the formula for calculating force?", + "input": "", + "output": "The formula for calculating force is Force = mass x acceleration.", + "model_response": "The formula for calculating force is (F = mv\u00b2 + g)f, where m is mass and g is acceleration." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "The company's innovative approach set it apart from its competitors.", + "model_response": "She was very innovative and always had a new idea." + }, + { + "instruction": "Generate a sentence using the word 'generous'.", + "input": "", + "output": "He is very generous and always helps those in need.", + "model_response": "She was very generous and gave the money to the poor." + }, + { + "instruction": "What is an antonym of 'sharp'?", + "input": "", + "output": "An antonym of 'sharp' is 'dull'.", + "model_response": "An antonym of 'sharp' is 'soft'." + }, + { + "instruction": "Describe the basic structure of a neuron.", + "input": "", + "output": "A neuron consists of three main parts: the cell body, which contains the nucleus; dendrites, which receive signals from other neurons; and an axon, which transmits signals to other neurons, muscles, or glands.", + "model_response": "A neuron is a type of synapse in the brain." + }, + { + "instruction": "Translate 'see you later' into Spanish.", + "input": "", + "output": "Hasta luego", + "model_response": "Est\u00e1n leyendo see you later." + }, + { + "instruction": "What are the first 10 elements of the Fibonacci sequence?", + "input": "", + "output": "0, 1, 1, 2, 3, 5, 8, 13, 21, 34.", + "model_response": "The first 10 elements of the Fibonacci sequence are: 1, 3, 5, 7, 11, 13, 15, 20." + }, + { + "instruction": "What is an antonym of 'transparent'?", + "input": "", + "output": "An antonym of 'transparent' is 'opaque'.", + "model_response": "An antonym of 'transparent' is 'illuminate'." + }, + { + "instruction": "What is the past tense of 'think'?", + "input": "", + "output": "The past tense of 'think' is 'thought'.", + "model_response": "The past tense of 'think' is 'to think'." + }, + { + "instruction": "Classify each sentence as either declarative, interrogative, imperative, or exclamatory.", + "input": "Please open the door.", + "output": "The classification of the sentence 'Please open the door.' is imperative.", + "model_response": "Please open the door." + }, + { + "instruction": "Rewrite the sentence to use a negative adverb.", + "input": "She always remembers to call.", + "output": "She never forgets to call.", + "model_response": "She always remembers to call." + }, + { + "instruction": "Convert 50 miles per hour to kilometers per hour.", + "input": "", + "output": "50 miles per hour is approximately 80.47 kilometers per hour.", + "model_response": "50 miles per hour is approximately 32.5 kilometers per hour." + } +] \ No newline at end of file diff --git a/Codes/ch07/01_main-chapter-code/instruction-data.json b/Codes/ch07/01_main-chapter-code/instruction-data.json new file mode 100644 index 0000000..ffda728 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/instruction-data.json @@ -0,0 +1,5502 @@ +[ + { + "instruction": "Evaluate the following phrase by transforming it into the spelling given.", + "input": "freind --> friend", + "output": "The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\"." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "He go to the park every day.", + "output": "He goes to the park every day." + }, + { + "instruction": "Convert 45 kilometers to meters.", + "input": "", + "output": "45 kilometers is 45000 meters." + }, + { + "instruction": "Rewrite this sentence to start with 'Although': Despite the rain, they went for a walk.", + "input": "", + "output": "Although it was raining, they went for a walk." + }, + { + "instruction": "What are the first 10 square numbers?", + "input": "", + "output": "1, 4, 9, 16, 25, 36, 49, 64, 81, 100." + }, + { + "instruction": "Suggest a more formal synonym for \"happy.\"", + "input": "", + "output": "A more formal synonym for \"happy\" is \"content.\"" + }, + { + "instruction": "Translate the following sentence into French.", + "input": "Where is the nearest restaurant?", + "output": "O\u00f9 est le restaurant le plus proche?" + }, + { + "instruction": "Look up the melting point of iron.", + "input": "", + "output": "The melting point of iron is approximately 1538 degrees Celsius or 2800 degrees Fahrenheit." + }, + { + "instruction": "Rewrite the following sentence so that it is in active voice.", + "input": "The cake was baked by Sarah.", + "output": "Sarah baked the cake." + }, + { + "instruction": "Convert the active sentence to passive: 'The manager approved the project.'", + "input": "", + "output": "The project was approved by the manager." + }, + { + "instruction": "What is the contraction for \"will not\"?", + "input": "", + "output": "The contraction for \"will not\" is \"won't\"." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "The actual fact is that he was late.", + "output": "The fact is that he was late." + }, + { + "instruction": "What is the capital of Indonesia?", + "input": "", + "output": "The capital of Indonesia is Jakarta." + }, + { + "instruction": "Provide a synonym for 'bright'.", + "input": "", + "output": "A synonym for 'bright' is 'radiant'." + }, + { + "instruction": "Suggest an alternate word for 'happy'.", + "input": "", + "output": "joyful." + }, + { + "instruction": "Name the device that measures atmospheric pressure.", + "input": "", + "output": "The device that measures atmospheric pressure is a barometer." + }, + { + "instruction": "List two antonyms for 'intelligent.'", + "input": "", + "output": "Two antonyms for 'intelligent' are 'stupid' and 'dumb.'" + }, + { + "instruction": "What is the pH of a neutral solution at 25\u00b0C?", + "input": "", + "output": "The pH of a neutral solution at 25\u00b0C is 7." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor", + "input": "'Her eyes are like stars.'", + "output": "Her eyes are stars in the night sky." + }, + { + "instruction": "Name three types of biomes.", + "input": "", + "output": "Forest, desert, and grassland." + }, + { + "instruction": "Identify the climax in 'Romeo and Juliet'.", + "input": "", + "output": "The climax of 'Romeo and Juliet' occurs when Romeo kills Tybalt, leading to his banishment and the subsequent tragic events." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The dog is very loyal.", + "output": "The dog is as loyal as a soldier." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "The storm started suddenly.", + "output": "The storm came out of the blue." + }, + { + "instruction": "Convert 10 feet to meters.", + "input": "", + "output": "10 feet is equal to approximately 3.048 meters." + }, + { + "instruction": "Convert this sentence to passive voice", + "input": "The chef cooked a delicious meal.", + "output": "A delicious meal was cooked by the chef." + }, + { + "instruction": "What is the plural form of 'mouse'?", + "input": "", + "output": "The plural form of 'mouse' is 'mice'." + }, + { + "instruction": "What is the opposite of 'thin'?", + "input": "", + "output": "The opposite of 'thin' is 'thick'." + }, + { + "instruction": "What is the state capital of California?", + "input": "", + "output": "The state capital of California is Sacramento." + }, + { + "instruction": "Translate 'family' into Italian.", + "input": "", + "output": "The Italian word for 'family' is 'famiglia'." + }, + { + "instruction": "What is the past tense of 'draw'?", + "input": "", + "output": "The past tense of 'draw' is 'drew.'" + }, + { + "instruction": "Alter the content of the sentence to use the past tense.", + "input": "The ship sails at dawn.", + "output": "The ship sailed at dawn." + }, + { + "instruction": "Generate a sentence that follows the pattern: 'Rarely do I _____'", + "input": "", + "output": "Rarely do I eat fast food." + }, + { + "instruction": "What is the atomic number of oxygen?", + "input": "", + "output": "The atomic number of oxygen is 8." + }, + { + "instruction": "What is a synonym of 'luminous'?", + "input": "", + "output": "A synonym for 'luminous' is 'bright'." + }, + { + "instruction": "What is the capital of Brazil?", + "input": "", + "output": "The capital of Brazil is Bras\u00edlia." + }, + { + "instruction": "List the ingredients required to make a basic salad.", + "input": "", + "output": "Ingredients for a basic salad typically include lettuce, tomatoes, cucumbers, and dressing." + }, + { + "instruction": "Rewrite the sentence using a synonym.", + "input": "It's hot outside.", + "output": "It's scorching outside." + }, + { + "instruction": "Provide a synonym for 'happy'.", + "input": "", + "output": "A synonym for 'happy' is 'content'." + }, + { + "instruction": "What is the derivative of x^3 + 5x^2 - 3x + 7 with respect to x?", + "input": "", + "output": "3x^2 + 10x - 3." + }, + { + "instruction": "Identify the point of view used in the sentence.", + "input": "She felt a sense of relief.", + "output": "The point of view used in the sentence is third person." + }, + { + "instruction": "Generate a sentence using the word 'gregarious'.", + "input": "", + "output": "He was gregarious, making friends wherever he went." + }, + { + "instruction": "What is the atomic number of carbon?", + "input": "", + "output": "The atomic number of carbon is 6." + }, + { + "instruction": "What is the formula for area of a triangle?", + "input": "", + "output": "The formula for the area of a triangle is 1/2 * base * height." + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Zebra, Elephant, Crocodile", + "output": "Crocodile, Elephant, Zebra" + }, + { + "instruction": "Reverse the order of the given phrase.", + "input": "Moon and stars", + "output": "Stars and moon" + }, + { + "instruction": "What is a synonym for 'begin'?", + "input": "", + "output": "A synonym for 'begin' is 'commence'." + }, + { + "instruction": "Find the next 3 terms in the following sequence: 4, 8, 12, 16.", + "input": "", + "output": "The next three terms in the sequence are: 20, 24, 28." + }, + { + "instruction": "What is the melting point of aluminum in Celsius?", + "input": "", + "output": "The melting point of aluminum is 660.3 degrees Celsius." + }, + { + "instruction": "What is an antonym of 'advanced'?", + "input": "", + "output": "An antonym of 'advanced' is 'primitive'." + }, + { + "instruction": "Name an antonym for 'happiness.'", + "input": "", + "output": "An antonym for 'happiness' is 'sadness.'" + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Ocassion", + "output": "The correct spelling is 'Occasion.'" + }, + { + "instruction": "What type of figurative language is used in 'She is the apple of my eye'?", + "input": "", + "output": "The figurative language used is a metaphor." + }, + { + "instruction": "Correct the spelling error in the sentence.", + "input": "I will atend the meeting tomorrow.", + "output": "The correct spelling is 'attend', not 'atend'." + }, + { + "instruction": "Arrange the following events in chronological order: World War II, French Revolution, Invention of the Telephone.", + "input": "", + "output": "1. French Revolution\n2. Invention of the Telephone\n3. World War II" + }, + { + "instruction": "What is the normal body temperature in Celsius?", + "input": "", + "output": "The normal body temperature is approximately 37 degrees Celsius." + }, + { + "instruction": "What is the cube of 4?", + "input": "", + "output": "The cube of 4 is 64." + }, + { + "instruction": "Classify the following into vegetables, animals, and metals.", + "input": "Broccoli, Iron, Elephant", + "output": "Vegetables: Broccoli\nMetals: Iron\nAnimals: Elephant" + }, + { + "instruction": "Translate 'I am learning Spanish' into Spanish.", + "input": "", + "output": "Estoy aprendiendo espa\u00f1ol." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "He response to the issue was quick.", + "output": "His response to the issue was quick." + }, + { + "instruction": "What is the melting point of gold in Celsius?", + "input": "", + "output": "The melting point of gold is 1064 degrees Celsius." + }, + { + "instruction": "Provide a synonym for 'fast'.", + "input": "", + "output": "A synonym for 'fast' is 'quick'." + }, + { + "instruction": "Translate 'What is your name?' into French.", + "input": "", + "output": "The French translation of 'What is your name?' is 'Comment vous appelez-vous?'." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very graceful.", + "output": "She is as graceful as a swan." + }, + { + "instruction": "What is an antonym of 'soft'?", + "input": "", + "output": "hard." + }, + { + "instruction": "What are the first 5 multiples of 4?", + "input": "", + "output": "4, 8, 12, 16, 20." + }, + { + "instruction": "What is an antonym of 'happy'?", + "input": "", + "output": "An antonym of 'happy' is 'sad.'" + }, + { + "instruction": "What type of galaxy is the Andromeda Galaxy?", + "input": "", + "output": "The Andromeda Galaxy is a spiral galaxy." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "She is very punctual.", + "output": "She is always on time." + }, + { + "instruction": "Generate a sentence using the word 'inevitable'.", + "input": "", + "output": "The collapse of the company was inevitable due to poor management." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He runs very fast.", + "output": "He runs as fast as a cheetah." + }, + { + "instruction": "What is the chemical symbol for aluminum?", + "input": "", + "output": "Al." + }, + { + "instruction": "Name the process by which water changes from liquid to gas.", + "input": "", + "output": "The process by which water changes from liquid to gas is called evaporation." + }, + { + "instruction": "What is the melting point of gold?", + "input": "", + "output": "The melting point of gold is approximately 1064 degrees Celsius." + }, + { + "instruction": "Rewrite the sentence using a different adjective.", + "input": "She has a fast car.", + "output": "She has a speedy car." + }, + { + "instruction": "Add the correct verb to the sentence.", + "input": "The students _________ the quiz answers.", + "output": "The students discussed the quiz answers." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "He is very lucky.", + "output": "He has a stroke of luck." + }, + { + "instruction": "Calculate the volume of a cylinder with a radius of 4 cm and a height of 10 cm.", + "input": "", + "output": "The volume of the cylinder is approximately 502.65 cubic centimeters." + }, + { + "instruction": "Edit the following sentence to correct any grammar errors.", + "input": "They was waiting for the bus all day.", + "output": "They were waiting for the bus all day." + }, + { + "instruction": "Generate a sentence using the word 'serendipity.'", + "input": "", + "output": "Finding the hidden beach was pure serendipity." + }, + { + "instruction": "What type of bond is formed when one atom donates an electron to another atom?", + "input": "", + "output": "The type of bond formed when one atom donates an electron to another atom is an ionic bond." + }, + { + "instruction": "Create a palindrome using the word 'racecar'.", + "input": "", + "output": "A racecar." + }, + { + "instruction": "What is the square of 12?", + "input": "", + "output": "The square of 12 is 144." + }, + { + "instruction": "Insert the correct punctuation at the provided places in the following sentence.", + "input": "Where are you going", + "output": "Where are you going?" + }, + { + "instruction": "Provide a synonym for 'strong'.", + "input": "", + "output": "A synonym for 'strong' is 'powerful'." + }, + { + "instruction": "Name a mammal that lays eggs.", + "input": "", + "output": "A mammal that lays eggs is the platypus." + }, + { + "instruction": "Name three types of precipitation.", + "input": "", + "output": "Three types of precipitation are rain, snow, and sleet." + }, + { + "instruction": "Name the process by which skin repairs itself after a cut.", + "input": "", + "output": "The process by which skin repairs itself after a cut is called regeneration." + }, + { + "instruction": "Generate a sentence using the word 'ephemeral'.", + "input": "", + "output": "The beauty of the sunset was ephemeral, fading from the sky within moments." + }, + { + "instruction": "Generate a sentence using the word 'harmony'.", + "input": "", + "output": "The choir's voices blended in perfect harmony." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The storm is very strong.", + "output": "The storm is a beast." + }, + { + "instruction": "What type of bond is formed when electrons are shared between atoms?", + "input": "", + "output": "The type of bond formed when electrons are shared between atoms is a covalent bond." + }, + { + "instruction": "Convert 3 meters to centimeters.", + "input": "", + "output": "3 meters is 300 centimeters." + }, + { + "instruction": "Arrange these technological inventions in chronological order: Telephone, Internet, Printing Press.", + "input": "", + "output": "1. Printing Press\n2. Telephone\n3. Internet" + }, + { + "instruction": "What is the main ingredient in traditional French baguettes?", + "input": "", + "output": "The main ingredient in traditional French baguettes is wheat flour." + }, + { + "instruction": "Classify the following text into either fiction or non-fiction.", + "input": "The documentary covers the impact of climate change on polar bears.", + "output": "Non-fiction." + }, + { + "instruction": "What is the boiling point of acetone in Celsius?", + "input": "", + "output": "The boiling point of acetone is 56 degrees Celsius." + }, + { + "instruction": "What is the capital of South Korea?", + "input": "", + "output": "The capital of South Korea is Seoul." + }, + { + "instruction": "Convert 300 seconds to minutes.", + "input": "", + "output": "300 seconds is equal to 5 minutes." + }, + { + "instruction": "Name a novel written by Jane Austen.", + "input": "", + "output": "One of the novels written by Jane Austen is 'Pride and Prejudice'." + }, + { + "instruction": "Provide a synonym for 'kind'.", + "input": "", + "output": "A synonym for 'kind' is 'benevolent'." + }, + { + "instruction": "Name a synonym for \"happiness.\"", + "input": "", + "output": "A synonym for \"happiness\" is \"joy.\"" + }, + { + "instruction": "Convert 8 kilometers to meters.", + "input": "", + "output": "8 kilometers is 8000 meters." + }, + { + "instruction": "Edit the sentence.", + "input": "She enjoys reading books.", + "output": "She enjoys reading books." + }, + { + "instruction": "Generate a question based on the following statement.", + "input": "Plants require sunlight to grow.", + "output": "Why do plants require sunlight to grow?" + }, + { + "instruction": "What is the capital of the Philippines?", + "input": "", + "output": "The capital of the Philippines is Manila." + }, + { + "instruction": "Name the process by which water cycles from the ground to the atmosphere and back.", + "input": "", + "output": "The process is called the water cycle." + }, + { + "instruction": "Define the term 'ecosystem'.", + "input": "", + "output": "An ecosystem is a community of interacting organisms and their physical environment." + }, + { + "instruction": "Convert the following verb to its past participle form: 'see'", + "input": "", + "output": "The past participle form of 'see' is 'seen'." + }, + { + "instruction": "What is the highest mountain in the world?", + "input": "", + "output": "The highest mountain in the world is Mount Everest." + }, + { + "instruction": "Convert the number 5 from decimal to binary.", + "input": "", + "output": "The binary equivalent of the decimal number 5 is 101." + }, + { + "instruction": "What is the chemical formula for carbonic acid?", + "input": "", + "output": "The chemical formula for carbonic acid is H2CO3." + }, + { + "instruction": "What is a synonym of 'optimistic'?", + "input": "", + "output": "A synonym for 'optimistic' is 'hopeful'." + }, + { + "instruction": "Change the sentence \"This should not have happened\" into a declarative sentence.", + "input": "", + "output": "This should have happened differently." + }, + { + "instruction": "Given a sentence, convert it into indirect speech.", + "input": "She said, \"I am tired.\"", + "output": "She said that she was tired." + }, + { + "instruction": "Classify this text as either a scientific article or a news report.", + "input": "The researchers found a new species of bird in the Amazon rainforest.", + "output": "Scientific article" + }, + { + "instruction": "Convert the following sentence to use a relative clause: The girl is a dancer. She lives next door.", + "input": "", + "output": "The girl who lives next door is a dancer." + }, + { + "instruction": "Classify the following words by their grammatical categories: run, happy, quickly", + "input": "", + "output": "Run: Verb\nHappy: Adjective\nQuickly: Adverb" + }, + { + "instruction": "Provide a synonym for 'brilliant'.", + "input": "", + "output": "A synonym for 'brilliant' is 'bright'." + }, + { + "instruction": "Put the following words in alphabetical order: lion, zebra, elephant.", + "input": "", + "output": "Elephant, lion, zebra." + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I hate this song!", + "output": "The sentiment of the sentence \"I hate this song!\" is negative." + }, + { + "instruction": "What type of literary device is used in 'The silence was a loud scream'?", + "input": "", + "output": "The literary device used is oxymoron." + }, + { + "instruction": "Rewrite the following phrase in the present tense.", + "input": "They had played soccer all afternoon.", + "output": "They play soccer all afternoon." + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I love this song!", + "output": "The sentiment of the sentence 'I love this song!' is positive. This is indicated by the word 'love' which shows strong liking or admiration towards the subject, in this case, the song." + }, + { + "instruction": "What is the capital of Canada?", + "input": "", + "output": "The capital of Canada is Ottawa." + }, + { + "instruction": "Edit the following sentence to remove all passive voice.", + "input": "The song was composed by the artist.", + "output": "The artist composed the song." + }, + { + "instruction": "Generate a sentence using the word 'determined'.", + "input": "", + "output": "She was determined to succeed despite the obstacles." + }, + { + "instruction": "Convert the active sentence to passive: 'Julia throws the ball.'", + "input": "", + "output": "The ball is thrown by Julia." + }, + { + "instruction": "Identify the predicate in the sentence.", + "input": "The quick brown fox jumps over the lazy dog.", + "output": "The predicate in the sentence is 'jumps over the lazy dog.'" + }, + { + "instruction": "What type of number is 0?", + "input": "", + "output": "0 is an integer." + }, + { + "instruction": "Take the following noun and turn it into a verb.", + "input": "Light", + "output": "To light." + }, + { + "instruction": "Translate 'Happy Birthday' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Happy Birthday' is 'Feliz Cumplea\u00f1os'." + }, + { + "instruction": "Translate 'Where is the library?' into Italian.", + "input": "", + "output": "Dove \u00e8 la biblioteca?" + }, + { + "instruction": "Rearrange the following list into descending order.", + "input": "12, 5, -3, 20, 9", + "output": "20, 12, 9, 5, -3" + }, + { + "instruction": "Generate a five-word phrase using the words \"moon,\" \"crisp,\" and \"night.\"", + "input": "", + "output": "\"Crisp moon illuminates the night.\"" + }, + { + "instruction": "What is a synonym of 'content'?", + "input": "", + "output": "A synonym for 'content' is 'satisfied.'" + }, + { + "instruction": "Rewrite the following sentence by adding one adverb.", + "input": "The cat slept on the mat.", + "output": "The cat peacefully slept on the mat." + }, + { + "instruction": "Convert 4 liters to milliliters.", + "input": "", + "output": "4 liters is 4000 milliliters." + }, + { + "instruction": "What is the atomic mass of carbon?", + "input": "", + "output": "The atomic mass of carbon is approximately 12.01 atomic mass units." + }, + { + "instruction": "Classify the following into animals, plants, and metals.", + "input": "Rose, gold, cat", + "output": "Animals: Cat\nPlants: Rose\nMetals: Gold" + }, + { + "instruction": "Classify the following numbers as even or odd.", + "input": "2, 7, 19, 20", + "output": "Even numbers: 2, 20\nOdd numbers: 7, 19" + }, + { + "instruction": "Translate the phrase 'Life is beautiful' into Italian.", + "input": "", + "output": "The Italian translation of 'Life is beautiful' is 'La vita \u00e8 bella'." + }, + { + "instruction": "What is the sum of 7 and 8?", + "input": "", + "output": "The sum of 7 and 8 is 15." + }, + { + "instruction": "Generate a random alphanumeric code of 10 characters.", + "input": "", + "output": "Here is a randomly generated 10-character alphanumeric code: 3F8G6H7JQ2." + }, + { + "instruction": "Provide a synonym for 'quiet'.", + "input": "", + "output": "A synonym for 'quiet' is 'silent'." + }, + { + "instruction": "Reword the sentence to use other words without changing the meaning.", + "input": "It rained heavily during the night.", + "output": "The night experienced significant rainfall." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "Her voice is very soothing.", + "output": "Her voice is a lullaby." + }, + { + "instruction": "Fill in the blanks with appropriate words.", + "input": "The sky was filled with colorful ________ at sunset.", + "output": "The sky was filled with colorful clouds at sunset." + }, + { + "instruction": "What is the boiling point of mercury in Celsius?", + "input": "", + "output": "The boiling point of mercury is 356.7 degrees Celsius." + }, + { + "instruction": "Change the following sentence to past perfect tense.", + "input": "They finish the game.", + "output": "They had finished the game." + }, + { + "instruction": "What is an antonym of 'bright'?", + "input": "", + "output": "An antonym of 'bright' is 'dark'." + }, + { + "instruction": "Calculate the area of a circle with a radius of 4 meters.", + "input": "", + "output": "The area of the circle is approximately 50.27 square meters." + }, + { + "instruction": "Determine the product of 7 and 8.", + "input": "", + "output": "The product of 7 and 8 is 56." + }, + { + "instruction": "Name the device used to measure humidity.", + "input": "", + "output": "The device used to measure humidity is a hygrometer." + }, + { + "instruction": "Identify the preposition in the sentence.", + "input": "The cat jumped over the fence.", + "output": "The preposition in the sentence is 'over'." + }, + { + "instruction": "Re-write the following sentence to use a different verb.", + "input": "I will help you with your homework.", + "output": "I will assist you with your homework." + }, + { + "instruction": "Identify the hypernym for the word 'rose.'", + "input": "", + "output": "The hypernym for the word 'rose' is 'flower.'" + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "He is very intelligent.", + "output": "He is a genius." + }, + { + "instruction": "What is the formula for calculating the volume of a cylinder?", + "input": "", + "output": "The formula for calculating the volume of a cylinder is \u03c0r\u00b2h, where r is the radius and h is the height." + }, + { + "instruction": "What is an antonym for 'difficult'?", + "input": "", + "output": "An antonym for 'difficult' is 'easy'." + }, + { + "instruction": "Generate a sentence using the word 'brilliant'.", + "input": "", + "output": "The scientist had a brilliant idea that changed the world." + }, + { + "instruction": "Automatically correct the given sentence.", + "input": "I want to go to london.", + "output": "I want to go to London." + }, + { + "instruction": "Convert this declarative sentence into an interrogative sentence.", + "input": "You need help.", + "output": "Do you need help?" + }, + { + "instruction": "Convert the sentence into a negative statement.", + "input": "You will help me.", + "output": "You will not help me." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "She response to the question was quick.", + "output": "Her response to the question was quick." + }, + { + "instruction": "What is the boiling point of oxygen in Celsius?", + "input": "", + "output": "The boiling point of oxygen is -183 degrees Celsius." + }, + { + "instruction": "Translate the word 'hello' into Spanish.", + "input": "", + "output": "The Spanish translation of 'hello' is 'hola'." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Orange", + "output": "The fruit 'Orange' is typically classified as a citrus fruit." + }, + { + "instruction": "Calculate the perimeter of a triangle with sides measuring 3, 4, and 5 cm.", + "input": "", + "output": "The perimeter of the triangle is 12 cm." + }, + { + "instruction": "Convert the sentence to use an indefinite pronoun.", + "input": "Someone left a message.", + "output": "A message was left by someone." + }, + { + "instruction": "What is the prime factorization of 210?", + "input": "", + "output": "2 \u00d7 3 \u00d7 5 \u00d7 7." + }, + { + "instruction": "Convert this active sentence to passive: 'The company employs ten workers.'", + "input": "", + "output": "Ten workers are employed by the company." + }, + { + "instruction": "Transform the given statement into an imperative sentence.", + "input": "I would like you to turn off the light.", + "output": "Please turn off the light." + }, + { + "instruction": "Generate three verbs that mean the same as 'to speak.'", + "input": "", + "output": "1. To talk\n2. To converse\n3. To chat" + }, + { + "instruction": "Edit the sentence below to remove any redundant words and make it more concise.", + "input": "I actually think that", + "output": "I think." + }, + { + "instruction": "Fill in the blank to complete the sentence.", + "input": "The most popular fruit in the world is the ___", + "output": "The most popular fruit in the world is the banana." + }, + { + "instruction": "Cut the following sentence into two, keeping the original meaning intact.", + "input": "She started her homework but was soon distracted by a phone call.", + "output": "She started her homework. However, she was soon distracted by a phone call." + }, + { + "instruction": "Generate a sentence using the word 'melancholy'.", + "input": "", + "output": "The melancholy tune brought tears to her eyes." + }, + { + "instruction": "Generate three adjectives to describe a river.", + "input": "", + "output": "1. Serene\n2. Flowing\n3. Winding" + }, + { + "instruction": "Name the longest river in the world.", + "input": "", + "output": "The longest river in the world is the Nile River." + }, + { + "instruction": "Translate 'Happy New Year' into Italian.", + "input": "", + "output": "Buon Anno Nuovo" + }, + { + "instruction": "Select the noun phrase in the following sentence.", + "input": "He enjoyed a delicious cake at the party.", + "output": "The noun phrases in the given sentence are \"a delicious cake\" and \"the party.\"" + }, + { + "instruction": "Summarize the plot of Shakespeare's play 'Macbeth'.", + "input": "", + "output": "Macbeth, a Scottish general, is led to wicked thoughts by the prophecies of three witches, leading him to murder the king and take the throne. His lust for power causes a descent into madness and destruction." + }, + { + "instruction": "Classify the following items: gold ring, daisy, python.", + "input": "", + "output": "Animals: Python\nPlants: Daisy\nMinerals: Gold ring" + }, + { + "instruction": "What is the capital of Greece?", + "input": "", + "output": "The capital of Greece is Athens." + }, + { + "instruction": "Convert the following number from decimal to hexadecimal: 255.", + "input": "", + "output": "The hexadecimal equivalent of the decimal number 255 is FF." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The music is very loud.", + "output": "The music is as loud as thunder." + }, + { + "instruction": "What is the capital of Argentina?", + "input": "", + "output": "The capital of Argentina is Buenos Aires." + }, + { + "instruction": "Name a punctuation mark used to indicate a question.", + "input": "", + "output": "A punctuation mark used to indicate a question is the question mark (?)." + }, + { + "instruction": "What is an antonym of \"ancient\"?", + "input": "", + "output": "An antonym of \"ancient\" is \"modern.\"" + }, + { + "instruction": "Determine the mass of an object that has a density of 10 g/cm\u00b3 and a volume of 2 cm\u00b3.", + "input": "", + "output": "The mass of the object is 20 grams." + }, + { + "instruction": "Convert this compound sentence to a complex sentence.", + "input": "She likes coffee, and he likes tea.", + "output": "She likes coffee while he likes tea." + }, + { + "instruction": "Identify the antagonist in 'Harry Potter and the Sorcerer's Stone'.", + "input": "", + "output": "The antagonist in 'Harry Potter and the Sorcerer's Stone' is Voldemort." + }, + { + "instruction": "What is the capital of Sweden?", + "input": "", + "output": "The capital of Sweden is Stockholm." + }, + { + "instruction": "Generate a sentence using the word 'curious'.", + "input": "", + "output": "The cat was curious about the new toy." + }, + { + "instruction": "Convert the given sentence into passive voice.", + "input": "The company will announce the results tomorrow.", + "output": "The results will be announced by the company tomorrow." + }, + { + "instruction": "Classify an input string as either a noun or a verb.", + "input": "jump", + "output": "\"jump\" can be classified as a verb." + }, + { + "instruction": "Split the following sentence into two declarative sentences: \"The book is interesting but the cover is dull\".", + "input": "", + "output": "The book is interesting.\nThe cover is dull." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The book is very boring.", + "output": "The book is a cure for insomnia." + }, + { + "instruction": "What is an antonym of 'fast'?", + "input": "", + "output": "slow." + }, + { + "instruction": "Calculate the kinetic energy of an object with a mass of 10 kilograms moving at a velocity of 5 meters per second.", + "input": "", + "output": "The kinetic energy of the object is 125 joules, calculated using the formula KE = 0.5 * mass * velocity^2." + }, + { + "instruction": "What is the chemical formula for ammonia?", + "input": "", + "output": "The chemical formula for ammonia is NH3." + }, + { + "instruction": "What is the boiling point of methane in Celsius?", + "input": "", + "output": "The boiling point of methane is -161.5 degrees Celsius." + }, + { + "instruction": "What is the formula for the area of a parallelogram?", + "input": "", + "output": "The formula for the area of a parallelogram is base \u00d7 height." + }, + { + "instruction": "What is the unit for measuring electrical resistance?", + "input": "", + "output": "The unit for measuring electrical resistance is the ohm." + }, + { + "instruction": "Given a sentence, convert it into direct speech.", + "input": "She said she was tired.", + "output": "'I am tired,' she said." + }, + { + "instruction": "Generate a sentence using the word 'gregarious.'", + "input": "", + "output": "He was known for his gregarious nature, always enjoying social gatherings." + }, + { + "instruction": "What is the past tense of 'choose'?", + "input": "", + "output": "chose." + }, + { + "instruction": "What is the melting point of platinum in Celsius?", + "input": "", + "output": "The melting point of platinum is 1768 degrees Celsius." + }, + { + "instruction": "Change this sentence grammatically.", + "input": "He have seen the movie.", + "output": "He has seen the movie." + }, + { + "instruction": "What is the capital of Russia?", + "input": "", + "output": "The capital of Russia is Moscow." + }, + { + "instruction": "What is the chemical symbol for potassium?", + "input": "", + "output": "The chemical symbol for potassium is K." + }, + { + "instruction": "Name three states of matter.", + "input": "", + "output": "The three states of matter are solid, liquid, and gas." + }, + { + "instruction": "What is an antonym of 'simplify'?", + "input": "", + "output": "An antonym of 'simplify' is 'complicate.'" + }, + { + "instruction": "List the ingredients necessary to bake a simple chocolate cake.", + "input": "", + "output": "Flour, sugar, cocoa powder, baking powder, eggs, milk, and oil." + }, + { + "instruction": "Classify the following species as reptiles or amphibians.", + "input": "Frog", + "output": "Frog is an amphibian." + }, + { + "instruction": "Name the process by which plants lose water vapor through their leaves.", + "input": "", + "output": "The process by which plants lose water vapor through their leaves is called transpiration." + }, + { + "instruction": "Classify the following items: diamond ring, oak tree, lion.", + "input": "", + "output": "Animals: Lion\nPlants: Oak tree\nMinerals: Diamond ring" + }, + { + "instruction": "Rewrite this sentence to eliminate the passive voice.", + "input": "The cake was made by Susan.", + "output": "Susan made the cake." + }, + { + "instruction": "What is the capital of the United Kingdom?", + "input": "", + "output": "The capital of the United Kingdom is London." + }, + { + "instruction": "Explain what a neutron star is.", + "input": "", + "output": "A neutron star is a type of stellar remnant that is extremely dense, composed mostly of neutrons, and is formed after a supernova explosion of a massive star." + }, + { + "instruction": "Name the process by which carbon dioxide is converted into organic compounds by plants.", + "input": "", + "output": "The process by which carbon dioxide is converted into organic compounds by plants is called photosynthesis." + }, + { + "instruction": "Edit the sentence to make it grammatically correct.", + "input": "He is more happier now.", + "output": "He is happier now." + }, + { + "instruction": "Name the process by which birds migrate south for the winter.", + "input": "", + "output": "The process by which birds migrate south for the winter is called migration." + }, + { + "instruction": "Generate a sentence using the word 'ambitious'.", + "input": "", + "output": "She is very ambitious and always sets high goals for herself." + }, + { + "instruction": "Describe the process of diffusion.", + "input": "", + "output": "Diffusion is the movement of particles from an area of higher concentration to an area of lower concentration until they are evenly distributed. It occurs in liquids and gases due to the random movement of particles." + }, + { + "instruction": "What is the chemical formula for potassium chloride?", + "input": "", + "output": "The chemical formula for potassium chloride is KCl." + }, + { + "instruction": "Convert the sentence into a question by adding a question tag.", + "input": "You are coming to the party.", + "output": "You are coming to the party, aren't you?" + }, + { + "instruction": "Translate 'My name is John' into Spanish.", + "input": "", + "output": "Mi nombre es John." + }, + { + "instruction": "Generate a sentence using the word 'mysterious'.", + "input": "", + "output": "The old mansion had a mysterious aura that intrigued everyone." + }, + { + "instruction": "What is the formula for the circumference of a circle?", + "input": "", + "output": "The formula for the circumference of a circle is 2\u03c0r, where r is the radius." + }, + { + "instruction": "Explain the water cycle in simple terms.", + "input": "", + "output": "The water cycle describes how water evaporates from the surface of the earth, rises into the atmosphere, cools and condenses into rain or snow in clouds, and falls again to the surface as precipitation." + }, + { + "instruction": "Convert 6 kilometers to meters.", + "input": "", + "output": "6 kilometers is 6000 meters." + }, + { + "instruction": "Translate 'He likes to play football' into French.", + "input": "", + "output": "Il aime jouer au football." + }, + { + "instruction": "What is the speed of sound in air?", + "input": "", + "output": "The speed of sound in air is approximately 343 meters per second." + }, + { + "instruction": "What is the contraction for 'it is'?", + "input": "", + "output": "The contraction for 'it is' is 'it's.'" + }, + { + "instruction": "Translate 'What are you doing?' into Italian.", + "input": "", + "output": "Cosa stai facendo?" + }, + { + "instruction": "Translate 'How old are you?' into German.", + "input": "", + "output": "The German translation of 'How old are you?' is 'Wie alt bist du?'." + }, + { + "instruction": "Convert 50 kilometers to miles.", + "input": "", + "output": "50 kilometers is approximately 31.07 miles." + }, + { + "instruction": "Change the following sentence to present perfect tense.", + "input": "They finish the game.", + "output": "They have finished the game." + }, + { + "instruction": "Identify the preposition in the sentence.", + "input": "The cat is under the table.", + "output": "The preposition in the sentence is 'under'." + }, + { + "instruction": "Name the device used to measure wind speed.", + "input": "", + "output": "The device used to measure wind speed is an anemometer." + }, + { + "instruction": "What is the symbol for the chemical element silver?", + "input": "", + "output": "The symbol for the chemical element silver is Ag." + }, + { + "instruction": "What is the opposite of 'large'?", + "input": "", + "output": "The opposite of 'large' is 'small'." + }, + { + "instruction": "What is the opposite of 'dark'?", + "input": "", + "output": "The opposite of 'dark' is 'light'." + }, + { + "instruction": "Convert the following verb to its past participle form: 'go'", + "input": "go", + "output": "The past participle form of 'go' is 'gone'." + }, + { + "instruction": "Rewrite this sentence to eliminate the passive voice.", + "input": "The book was written by the author.", + "output": "The author wrote the book." + }, + { + "instruction": "Suggest a continuation for the sentence.", + "input": "Once the sun rose,", + "output": "'the city came to life.'" + }, + { + "instruction": "Calculate the perimeter of a hexagon with side lengths of 4 cm.", + "input": "", + "output": "The perimeter of the hexagon is 24 cm." + }, + { + "instruction": "Rewrite the sentence using alliteration.", + "input": "The cat chased the mouse.", + "output": "The cunning cat chased the mouse." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very brave.", + "output": "He is as brave as a lion." + }, + { + "instruction": "Translate 'Excuse me' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Excuse me' is 'Perd\u00f3n' or 'Disculpe'." + }, + { + "instruction": "What is an antonym of 'begin'?", + "input": "", + "output": "An antonym of 'begin' is 'end'." + }, + { + "instruction": "Create a sentence that begins with 'If only,'", + "input": "", + "output": "If only I had studied harder..." + }, + { + "instruction": "Identify the literary term used in 'It's raining cats and dogs.'", + "input": "", + "output": "The literary term used is an idiom." + }, + { + "instruction": "Calculate the area of a circle with a radius of 5 meters.", + "input": "", + "output": "The area of the circle is approximately 78.54 square meters." + }, + { + "instruction": "Given the following sentence, rewrite it by removing the passive form", + "input": "The final decision was made by the manager.", + "output": "The manager made the final decision." + }, + { + "instruction": "Name the process by which plants convert carbon dioxide and water into oxygen and glucose.", + "input": "", + "output": "The process by which plants convert carbon dioxide and water into oxygen and glucose is called photosynthesis." + }, + { + "instruction": "Rewrite the sentence to use a simile.", + "input": "She is very calm under pressure.", + "output": "She is as calm as a cucumber under pressure." + }, + { + "instruction": "What type of animal is an octopus?", + "input": "", + "output": "An octopus is a cephalopod mollusk." + }, + { + "instruction": "Provide a word that rhymes with \"care.\"", + "input": "", + "output": "A word that rhymes with \"care\" is \"fare.\"" + }, + { + "instruction": "What is a synonym for 'small'?", + "input": "", + "output": "A synonym for 'small' is 'little'." + }, + { + "instruction": "Convert the active sentence to passive: 'The gardener waters the plants daily.'", + "input": "", + "output": "The plants are watered daily by the gardener." + }, + { + "instruction": "Rewrite the following phrase as a question.", + "input": "We must finish the project by tomorrow.", + "output": "Must we finish the project by tomorrow?" + }, + { + "instruction": "List the colors of the rainbow.", + "input": "", + "output": "The colors of the rainbow are red, orange, yellow, green, blue, indigo, violet." + }, + { + "instruction": "Look up the freezing point of water.", + "input": "", + "output": "The freezing point of water is 0 degrees Celsius or 32 degrees Fahrenheit." + }, + { + "instruction": "Arrange the given numbers in descending order.", + "input": "5, 12, 8, 3, 15", + "output": "15, 12, 8, 5, 3." + }, + { + "instruction": "Translate 'I love reading books' into German.", + "input": "", + "output": "The German translation of 'I love reading books' is 'Ich liebe es, B\u00fccher zu lesen.'" + }, + { + "instruction": "What is a synonym for 'large'?", + "input": "", + "output": "A synonym for 'large' is 'big'." + }, + { + "instruction": "Classify the following species as dogs or cats.", + "input": "Siamese", + "output": "Siamese is a cat." + }, + { + "instruction": "Generate a sentence using the word 'exuberant'.", + "input": "", + "output": "The children were exuberant as they ran through the park." + }, + { + "instruction": "Generate a sentence using the word 'temporary.'", + "input": "", + "output": "Her stay at the hotel was only temporary." + }, + { + "instruction": "What is the formula for converting Celsius to Kelvin?", + "input": "", + "output": "The formula for converting Celsius to Kelvin is K = C + 273.15." + }, + { + "instruction": "Sort the following numbers in ascending order.", + "input": "7, 1, 8, 4, 3.", + "output": "The numbers in ascending order would be: 1, 3, 4, 7, 8." + }, + { + "instruction": "Categorize the given list of animals.", + "input": "Dog, Eagle, Trout", + "output": "Mammals: Dog\nBirds: Eagle\nFish: Trout" + }, + { + "instruction": "What is an antonym for 'success'?", + "input": "", + "output": "An antonym for 'success' is 'failure'." + }, + { + "instruction": "What is an antonym of 'brave'?", + "input": "", + "output": "An antonym of 'brave' is 'cowardly'." + }, + { + "instruction": "Change this sentence grammatically.", + "input": "They has seen the movie.", + "output": "They have seen the movie." + }, + { + "instruction": "Transform the following sentence into the passive voice.", + "input": "They will build a new bridge.", + "output": "A new bridge will be built by them." + }, + { + "instruction": "Generate a sentence using the word 'enthusiastic'.", + "input": "", + "output": "She is enthusiastic about learning new things." + }, + { + "instruction": "Transform the following sentence into a question using 'should.'", + "input": "You need to see a doctor.", + "output": "Should you see a doctor?" + }, + { + "instruction": "Explain what a limerick is.", + "input": "", + "output": "A limerick is a humorous five-line poem with a specific rhythm and rhyme scheme, typically AABBA." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'attractive'." + }, + { + "instruction": "What is the boiling point of mercury in Fahrenheit?", + "input": "", + "output": "The boiling point of mercury is 674 degrees Fahrenheit." + }, + { + "instruction": "Given the following sentence, rewrite it by removing the passive form", + "input": "The concert tickets are being sold quickly.", + "output": "They are selling the concert tickets quickly." + }, + { + "instruction": "Name the device that measures temperature.", + "input": "", + "output": "The device that measures temperature is a thermometer." + }, + { + "instruction": "Edit the following sentence to correct any grammar errors.", + "input": "They was trying to fix the car all day.", + "output": "They were trying to fix the car all day." + }, + { + "instruction": "What is the melting point of lead in Celsius?", + "input": "", + "output": "The melting point of lead is 327.5 degrees Celsius." + }, + { + "instruction": "Change the sentences from active voice to passive voice.", + "input": "The chef will prepare the meal.", + "output": "The meal will be prepared by the chef." + }, + { + "instruction": "Convert the following statement into a question.", + "input": "She is going to the market.", + "output": "Is she going to the market?" + }, + { + "instruction": "Name three types of flowers.", + "input": "", + "output": "Three types of flowers are roses, lilies, and tulips." + }, + { + "instruction": "What is the primary function of the roots of a plant?", + "input": "", + "output": "The primary function of plant roots is to absorb water and nutrients from the soil. They also anchor the plant in place and store food and nutrients." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The computer is very slow.", + "output": "The computer is a snail." + }, + { + "instruction": "What is the boiling point of neon in Celsius?", + "input": "", + "output": "The boiling point of neon is -246 degrees Celsius." + }, + { + "instruction": "Name the process by which iron rusts.", + "input": "", + "output": "The process by which iron rusts is called oxidation." + }, + { + "instruction": "Convert 400 meters to kilometers.", + "input": "", + "output": "400 meters is 0.4 kilometers." + }, + { + "instruction": "Explain the process of erosion.", + "input": "", + "output": "Erosion is the process by which soil, rock, and other surface material are worn away and transported by natural forces such as wind or water. It can shape landscapes and affect ecosystems." + }, + { + "instruction": "Convert 50 centimeters to meters.", + "input": "", + "output": "50 centimeters is 0.5 meters." + }, + { + "instruction": "Identify the pronoun in the sentence.", + "input": "She decided to go to the store.", + "output": "The pronoun in the sentence is 'She'." + }, + { + "instruction": "Make the second sentence a dependent clause.", + "input": "She went to the market. She bought some fruit.", + "output": "She went to the market, where she bought some fruit." + }, + { + "instruction": "Correct the verb tense error in the sentence.", + "input": "She go to school every day.", + "output": "The corrected sentence should be: 'She goes to school every day.'" + }, + { + "instruction": "What is the capital of New Zealand?", + "input": "", + "output": "The capital of New Zealand is Wellington." + }, + { + "instruction": "Evaluate this sentence for spelling and grammar mistakes.", + "input": "I recieved an invitation to an exciting evenement.", + "output": "There are two errors in the sentence. The corrected sentence should be: \"I received an invitation to an exciting event.\"" + }, + { + "instruction": "Rewrite the sentence in future continuous tense.", + "input": "They will complete the project.", + "output": "They will be completing the project." + }, + { + "instruction": "Rewrite the following sentence in a more formal way:\n\"I need to find out what's wrong.\"", + "input": "", + "output": "I must ascertain what is incorrect." + }, + { + "instruction": "Convert 5 kilometers to meters.", + "input": "", + "output": "5 kilometers is 5000 meters." + }, + { + "instruction": "What is the chemical formula for carbon dioxide?", + "input": "", + "output": "The chemical formula for carbon dioxide is CO2." + }, + { + "instruction": "Generate a sentence using the word 'courage'.", + "input": "", + "output": "She showed great courage during the rescue mission." + }, + { + "instruction": "What is the opposite of 'old'?", + "input": "", + "output": "The opposite of 'old' is 'young'." + }, + { + "instruction": "Generate a sentence using the word 'resilient'.", + "input": "", + "output": "She is resilient and always bounces back from challenges." + }, + { + "instruction": "Rewrite the following sentence to include a hyperbole: 'I'm very tired.'", + "input": "I'm very tired.", + "output": "I'm so tired I could sleep for a year." + }, + { + "instruction": "What is the smallest planet in our solar system?", + "input": "", + "output": "The smallest planet in our solar system is Mercury." + }, + { + "instruction": "What is the symbol for gold on the periodic table?", + "input": "", + "output": "The symbol for gold on the periodic table is Au." + }, + { + "instruction": "Generate a sentence using the word 'meticulous'.", + "input": "", + "output": "She is very meticulous and always pays attention to details." + }, + { + "instruction": "What is the freezing point of ethanol in Celsius?", + "input": "", + "output": "The freezing point of ethanol is -114.1 degrees Celsius." + }, + { + "instruction": "Classify this sentence into one of the following topics: health, economy, technology, music", + "input": "Spotify has revolutionized how we listen to music.", + "output": "The topic of the given sentence is music." + }, + { + "instruction": "What is the chemical formula for sulfur dioxide?", + "input": "", + "output": "The chemical formula for sulfur dioxide is SO2." + }, + { + "instruction": "Translate the phrase 'I love you' into Italian.", + "input": "", + "output": "The Italian translation of 'I love you' is 'Ti amo'." + }, + { + "instruction": "Create a sentence using the word 'melancholy'.", + "input": "", + "output": "There was a melancholy tone in his voice as he recounted the story." + }, + { + "instruction": "Given a sentence, count the number of words.", + "input": "The quick brown fox jumps over the lazy dog.", + "output": "The sentence \"The quick brown fox jumps over the lazy dog.\" contains 9 words." + }, + { + "instruction": "Provide the past participle form of 'break'.", + "input": "", + "output": "The past participle form of 'break' is 'broken'." + }, + { + "instruction": "Generate a random five-letter word.", + "input": "", + "output": "quake." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "They goes to the market every Sunday.", + "output": "They go to the market every Sunday." + }, + { + "instruction": "Provide a synonym for 'amazing'.", + "input": "", + "output": "A synonym for 'amazing' is 'incredible'." + }, + { + "instruction": "Generate a sentence using the word 'determined'.", + "input": "", + "output": "She is determined to achieve her dreams despite the obstacles." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Banana", + "output": "The fruit 'Banana' is typically classified as a tropical fruit." + }, + { + "instruction": "Identify the main verb in the sentence.", + "input": "She danced gracefully.", + "output": "The main verb in the sentence is 'danced'." + }, + { + "instruction": "Provide a synonym for 'old'.", + "input": "", + "output": "A synonym for 'old' is 'ancient'." + }, + { + "instruction": "Describe the color red in 3 words.", + "input": "", + "output": "Bold, vibrant, intense." + }, + { + "instruction": "Translate 'I love books' into Spanish.", + "input": "", + "output": "Me encantan los libros." + }, + { + "instruction": "Name the process by which animals release energy from food.", + "input": "", + "output": "The process by which animals release energy from food is called respiration." + }, + { + "instruction": "What is the plural form of 'sheep'?", + "input": "", + "output": "The plural form of 'sheep' is 'sheep.'" + }, + { + "instruction": "Convert the following statement into a question.", + "input": "You were at the concert last night.", + "output": "Were you at the concert last night?" + }, + { + "instruction": "Edit the given text to ensure all plural nouns are spelled correctly.", + "input": "The boxes was stacked neatly.", + "output": "The boxes were stacked neatly." + }, + { + "instruction": "What is an antonym of 'external'?", + "input": "", + "output": "An antonym of 'external' is 'internal'." + }, + { + "instruction": "Identify the correct word to complete the sentence.", + "input": "She presented her ideas with _____.", + "output": "clarity" + }, + { + "instruction": "Edit the given sentence to make it grammatically correct.", + "input": "She dont have any money", + "output": "She doesn't have any money." + }, + { + "instruction": "Generate a sentence using the word 'compassionate'.", + "input": "", + "output": "She is compassionate and always helps others in need." + }, + { + "instruction": "Find a rhyming word for the word 'book.'", + "input": "", + "output": "A rhyming word for the word 'book' could be 'look.'" + }, + { + "instruction": "Name the process by which plants convert sunlight into energy.", + "input": "", + "output": "The process by which plants convert sunlight into energy is photosynthesis." + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She did not attend the meeting. She was ill.", + "output": "She did not attend the meeting because she was ill." + }, + { + "instruction": "What is the opposite of 'retain'?", + "input": "", + "output": "The opposite of 'retain' is 'release'." + }, + { + "instruction": "Translate 'Good luck' into Japanese.", + "input": "", + "output": "\u9811\u5f35\u3063\u3066 (Ganbatte)" + }, + { + "instruction": "List the ingredients required to make a sandwich.", + "input": "", + "output": "Ingredients for a sandwich typically include bread, meat or vegetables, cheese, and condiments." + }, + { + "instruction": "What is an antonym of 'broad'?", + "input": "", + "output": "An antonym of 'broad' is 'narrow'." + }, + { + "instruction": "Rewrite this statement as an imperative sentence.", + "input": "You should finish your assignment.", + "output": "Finish your assignment." + }, + { + "instruction": "What is an antonym for 'young'?", + "input": "", + "output": "An antonym for 'young' is 'old'." + }, + { + "instruction": "What is the symbol for the chemical element iron?", + "input": "", + "output": "The symbol for the chemical element iron is Fe." + }, + { + "instruction": "What is the abbreviation for 'corporation'?", + "input": "", + "output": "The abbreviation for 'corporation' is 'corp.'" + }, + { + "instruction": "Translate 'Please' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Please' is 'Por favor'." + }, + { + "instruction": "What is the chemical formula for sodium hydroxide?", + "input": "", + "output": "The chemical formula for sodium hydroxide is NaOH." + }, + { + "instruction": "Classify the following numbers as prime or composite.", + "input": "17, 22, 29.", + "output": "Prime numbers: 17, 29\nComposite numbers: 22" + }, + { + "instruction": "Provide the past tense form of the verb 'swim.'", + "input": "", + "output": "The past tense of the verb 'swim' is 'swam.'" + }, + { + "instruction": "What is the boiling point of gold in Celsius?", + "input": "", + "output": "The boiling point of gold is 2970 degrees Celsius." + }, + { + "instruction": "Reverse this array", + "input": "[10, 20, 30, 40, 50]", + "output": "Here is the reversed array: [50, 40, 30, 20, 10]" + }, + { + "instruction": "What is the melting point of silver in Celsius?", + "input": "", + "output": "The melting point of silver is 961.8 degrees Celsius." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "Her innovative ideas led to the success of the project." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "He spoke to the issue quickly.", + "output": "He quickly addressed the issue." + }, + { + "instruction": "What is the derivative of the function f(x) = 3x^2?", + "input": "", + "output": "The derivative of the function f(x) = 3x^2 is f'(x) = 6x." + }, + { + "instruction": "What type of galaxy is the Milky Way?", + "input": "", + "output": "The Milky Way is a barred spiral galaxy." + }, + { + "instruction": "What type of bond does sodium chloride (table salt) have?", + "input": "", + "output": "Sodium chloride (table salt) has an ionic bond." + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Definately", + "output": "The correct spelling is 'Definitely.'" + }, + { + "instruction": "Define the term 'personification'.", + "input": "", + "output": "Personification is a figure of speech where human qualities are given to animals, objects, or ideas." + }, + { + "instruction": "Identify the protagonist in the novel 'To Kill a Mockingbird'.", + "input": "", + "output": "The protagonist in 'To Kill a Mockingbird' is Scout Finch." + }, + { + "instruction": "What is the symbol for the chemical element gold?", + "input": "", + "output": "The symbol for the chemical element gold is Au." + }, + { + "instruction": "Rewrite the following sentence using passive voice.", + "input": "The chef prepared a delicious meal.", + "output": "A delicious meal was prepared by the chef." + }, + { + "instruction": "What is the chemical symbol for gold?", + "input": "", + "output": "The chemical symbol for gold is Au." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very fast.", + "output": "She is as fast as a cheetah." + }, + { + "instruction": "What is the opposite of the adjective 'generous'?", + "input": "", + "output": "The opposite of 'generous' is 'stingy' or 'miserly'." + }, + { + "instruction": "Identify the main subject in the sentence.", + "input": "The noisy children played in the park.", + "output": "The main subject in the sentence is \"The noisy children.\"" + }, + { + "instruction": "Capitalize each word in the sentence.", + "input": "i love watching the sunrise", + "output": "I Love Watching The Sunrise" + }, + { + "instruction": "What is the chemical formula for calcium chloride?", + "input": "", + "output": "The chemical formula for calcium chloride is CaCl2." + }, + { + "instruction": "What is the formula for the area of a trapezoid?", + "input": "", + "output": "The formula for the area of a trapezoid is 0.5 * (Base1 + Base2) * Height." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very tall.", + "output": "He is as tall as a giraffe." + }, + { + "instruction": "Explain the term 'ecosystem'.", + "input": "", + "output": "An ecosystem is a community of living organisms interacting with each other and their physical environment. It includes biotic (plants, animals, microbes) and abiotic (climate, soil, water) components." + }, + { + "instruction": "What is the capital of Egypt?", + "input": "", + "output": "The capital of Egypt is Cairo." + }, + { + "instruction": "What is the opposite of 'brave'?", + "input": "", + "output": "The opposite of 'brave' is 'cowardly'." + }, + { + "instruction": "Calculate the volume of a sphere with a radius of 3 cm.", + "input": "", + "output": "The volume of the sphere is approximately 113.10 cubic cm." + }, + { + "instruction": "Convert 2 kilometers to meters.", + "input": "", + "output": "2 kilometers is 2000 meters." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The garden is very beautiful.", + "output": "The garden is a paradise." + }, + { + "instruction": "What is the chemical symbol for oxygen?", + "input": "", + "output": "The chemical symbol for oxygen is O." + }, + { + "instruction": "What is the square root of 225?", + "input": "", + "output": "The square root of 225 is 15." + }, + { + "instruction": "What is the opposite of 'tall'?", + "input": "", + "output": "The opposite of 'tall' is 'short'." + }, + { + "instruction": "What is the hypotenuse of a right triangle with legs of 3 cm and 4 cm?", + "input": "", + "output": "The hypotenuse of the triangle is 5 cm." + }, + { + "instruction": "What is the capital of South Africa?", + "input": "", + "output": "The capital of South Africa is Pretoria." + }, + { + "instruction": "Provide an example of a compound sentence.", + "input": "", + "output": "I played the piano, and she sang along." + }, + { + "instruction": "What is the opposite of 'expand'?", + "input": "", + "output": "The opposite of 'expand' is 'contract'." + }, + { + "instruction": "Convert 20 miles per hour to kilometers per hour.", + "input": "", + "output": "20 miles per hour is approximately 32.19 kilometers per hour." + }, + { + "instruction": "Translate the sentence to French: 'How old are you?'", + "input": "", + "output": "Quel \u00e2ge avez-vous?" + }, + { + "instruction": "What is the opposite of 'early'?", + "input": "", + "output": "The opposite of 'early' is 'late'." + }, + { + "instruction": "Generate three verbs that mean the same as \"to run.\"", + "input": "", + "output": "1. To sprint\n2. To dash\n3. To jog" + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "She is very smart.", + "output": "She is a walking encyclopedia." + }, + { + "instruction": "Convert the following sentence to passive voice: 'The committee approved the new guidelines.'", + "input": "The committee approved the new guidelines.", + "output": "The new guidelines were approved by the committee." + }, + { + "instruction": "Classify this speech as either formal or informal.", + "input": "Could you please pass the salt?", + "output": "This speech is formal." + }, + { + "instruction": "What is the chemical formula for acetic acid?", + "input": "", + "output": "The chemical formula for acetic acid is CH3COOH." + }, + { + "instruction": "Determine the state of matter for oxygen at room temperature.", + "input": "", + "output": "Oxygen is in a gaseous state at room temperature." + }, + { + "instruction": "Provide a word that rhymes with 'day.'", + "input": "", + "output": "A word that rhymes with 'day' is 'play.'" + }, + { + "instruction": "Convert the following verb to its past participle form: 'begin'", + "input": "", + "output": "The past participle form of 'begin' is 'begun'." + }, + { + "instruction": "Combine the two ideas in the sentence.", + "input": "Learning a new language is useful. It can be challenging.", + "output": "Learning a new language is useful but can be challenging." + }, + { + "instruction": "What is the formula for density?", + "input": "", + "output": "The formula for density is density = mass/volume." + }, + { + "instruction": "What is the plural form of \"goose\"?", + "input": "", + "output": "The plural form of \"goose\" is \"geese.\"" + }, + { + "instruction": "Explain the process of convection.", + "input": "", + "output": "Convection is the transfer of heat through the movement of fluids (liquids or gases). It occurs when a warmer area of a liquid or gas rises and is replaced by a cooler area, creating a circulation pattern." + }, + { + "instruction": "Convert 2000 millimeters to meters.", + "input": "", + "output": "2000 millimeters is equal to 2 meters." + }, + { + "instruction": "Name an antonym for 'fear.'", + "input": "", + "output": "An antonym for 'fear' is 'courage.'" + }, + { + "instruction": "What is the opposite of 'high'?", + "input": "", + "output": "The opposite of 'high' is 'low'." + }, + { + "instruction": "What is the abbreviation for \"Doctor of Philosophy\"?", + "input": "", + "output": "The abbreviation for \"Doctor of Philosophy\" is Ph.D." + }, + { + "instruction": "What type of animal is a 'koala'?", + "input": "", + "output": "A koala is a marsupial." + }, + { + "instruction": "Pick out the correct verb from the following list.", + "input": "run, mountain, quickly", + "output": "The correct verb from the list is \"run.\"" + }, + { + "instruction": "Classify the given sentiment as either positive, negative, or neutral.", + "input": "This is a great opportunity!", + "output": "Positive" + }, + { + "instruction": "What is the formula for speed?", + "input": "", + "output": "The formula for speed is speed = distance/time." + }, + { + "instruction": "Name an antonym for 'big.'", + "input": "", + "output": "An antonym for 'big' is 'small.'" + }, + { + "instruction": "Name a solid commonly found in the Earth's crust.", + "input": "", + "output": "A solid commonly found in the Earth's crust is granite." + }, + { + "instruction": "What is the melting point of zinc in Celsius?", + "input": "", + "output": "The melting point of zinc is 419.5 degrees Celsius." + }, + { + "instruction": "List 5 different types of vegetables", + "input": "", + "output": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato (biologically a fruit, but often classified as a vegetable)\n5. Spinach" + }, + { + "instruction": "Classify the following substances as acid, base, or salt.", + "input": "Vinegar, Baking soda, Table salt", + "output": "Acid: Vinegar\nBase: Baking soda\nSalt: Table salt" + }, + { + "instruction": "Create an alliteration with the word 'moon'.", + "input": "", + "output": "The magnificent moon moved majestically across the midnight sky." + }, + { + "instruction": "What is the primary function of the cell membrane?", + "input": "", + "output": "The primary function of the cell membrane is to protect the cell by controlling the movement of substances in and out of the cell. It also provides structural support and facilitates communication between cells." + }, + { + "instruction": "What is the opposite of 'young'?", + "input": "", + "output": "The opposite of 'young' is 'old'." + }, + { + "instruction": "Determine the mass of an object that has a density of 8 g/cm\u00b3 and a volume of 3 cm\u00b3.", + "input": "", + "output": "The mass of the object is 24 grams." + }, + { + "instruction": "Provide the correct contraction for 'they are'.", + "input": "", + "output": "The contraction for 'they are' is 'they're'." + }, + { + "instruction": "Describe how a rainbow is formed.", + "input": "", + "output": "A rainbow is formed when sunlight is refracted, dispersed, and reflected in water droplets, resulting in a spectrum of light appearing in the sky." + }, + { + "instruction": "What is the plural form of 'octopus'?", + "input": "", + "output": "The plural form of 'octopus' can be 'octopuses' or 'octopi'." + }, + { + "instruction": "Rewrite the sentence using a metaphor: 'The assignment was a breeze.'", + "input": "The assignment was a breeze.", + "output": "The assignment was a piece of cake." + }, + { + "instruction": "Make the first sentence a dependent clause.", + "input": "I finished my work. I went home.", + "output": "After finishing my work, I went home." + }, + { + "instruction": "What is the boiling point of hydrogen?", + "input": "", + "output": "The boiling point of hydrogen is -252.9 degrees Celsius." + }, + { + "instruction": "Describe the process of evaporation.", + "input": "", + "output": "Evaporation is the process by which water changes from a liquid to a gas due to heat." + }, + { + "instruction": "Generate a sentence using the word 'illuminate'.", + "input": "", + "output": "The lamp helped to illuminate the dark room." + }, + { + "instruction": "What is the chemical formula for carbon monoxide?", + "input": "", + "output": "The chemical formula for carbon monoxide is CO." + }, + { + "instruction": "What is the atomic number of helium?", + "input": "", + "output": "The atomic number of helium is 2." + }, + { + "instruction": "Arrange the given numbers in ascending order.", + "input": "8, 3, 15, 1, 7", + "output": "1, 3, 7, 8, 15." + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Zebra, Elephant, Crocodile", + "output": "sorted in alphabetical order: Crocodile, Elephant, Zebra" + }, + { + "instruction": "Pick the best synonym for the given word.", + "input": "Gather", + "output": "Collect." + }, + { + "instruction": "What is the boiling point of nitrogen in Celsius?", + "input": "", + "output": "The boiling point of nitrogen is -196 degrees Celsius." + }, + { + "instruction": "Correct the misuse of 'your' in the sentence.", + "input": "Your going to love this place.", + "output": "The corrected sentence should be: 'You're going to love this place.'" + }, + { + "instruction": "Identify the type of triangle with angles of 90 degrees, 45 degrees, and 45 degrees.", + "input": "", + "output": "A triangle with angles of 90 degrees, 45 degrees, and 45 degrees is a right isosceles triangle." + }, + { + "instruction": "What is the chemical formula for sodium bicarbonate?", + "input": "", + "output": "The chemical formula for sodium bicarbonate is NaHCO3." + }, + { + "instruction": "What is the melting point of iron in Celsius?", + "input": "", + "output": "The melting point of iron is 1538 degrees Celsius." + }, + { + "instruction": "What is a synonym of 'meticulous'?", + "input": "", + "output": "A synonym for 'meticulous' is 'thorough'." + }, + { + "instruction": "What is a synonym of 'abundant'?", + "input": "", + "output": "A synonym for 'abundant' is 'plentiful'." + }, + { + "instruction": "List the ingredients to make pancakes.", + "input": "", + "output": "Ingredients for pancakes include flour, eggs, milk, and sugar." + }, + { + "instruction": "What is the primary function of red blood cells in the human body?", + "input": "", + "output": "The primary function of red blood cells is to carry oxygen from the lungs to the body's tissues and to return carbon dioxide from the tissues to the lungs." + }, + { + "instruction": "Translate 'How do you do?' into French.", + "input": "", + "output": "Comment allez-vous?" + }, + { + "instruction": "Construct a sentence using the words \"achieve\", \"goal\", and \"perseverance\".", + "input": "", + "output": "With perseverance, you can achieve any goal." + }, + { + "instruction": "Generate a five-word phrase using the words 'star,' 'night,' and 'bright.'", + "input": "", + "output": "'The bright star illuminated night.'" + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I hate this song!", + "output": "The sentiment of the sentence \"I hate this song!\" is negative. This is indicated by the word \"hate\" which shows strong aversion or dislike towards the subject, in this case, the song." + }, + { + "instruction": "Provide the past participle form of 'choose.'", + "input": "", + "output": "The past participle form of 'choose' is 'chosen.'" + }, + { + "instruction": "Rewrite the sentence so that it does not contain any passive voice.", + "input": "The artwork was displayed in the gallery.", + "output": "The gallery displayed the artwork." + }, + { + "instruction": "Pick out the verb from the following list.", + "input": "tree, sing, blue", + "output": "The correct verb from the list is 'sing.'" + }, + { + "instruction": "What are the first 10 odd numbers?", + "input": "", + "output": "1, 3, 5, 7, 9, 11, 13, 15, 17, 19." + }, + { + "instruction": "Generate a past-tense verb that describes a person singing.", + "input": "", + "output": "Sang." + }, + { + "instruction": "Find a rhyming word for the word \"care.\"", + "input": "", + "output": "A rhyming word for the word \"care\" could be \"bare.\"" + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She missed the bus. She walked to school.", + "output": "She missed the bus, so she walked to school." + }, + { + "instruction": "What is the boiling point of ethanol in Celsius?", + "input": "", + "output": "The boiling point of ethanol is 78 degrees Celsius." + }, + { + "instruction": "Convert the temperature from Celsius to Kelvin.", + "input": "25 degrees Celsius", + "output": "25 degrees Celsius is 298.15 Kelvin." + }, + { + "instruction": "What is an antonym for 'weak'?", + "input": "", + "output": "An antonym for 'weak' is 'strong.'" + }, + { + "instruction": "What is the opposite of 'soft'?", + "input": "", + "output": "The opposite of 'soft' is 'hard'." + }, + { + "instruction": "Name a synonym for 'exciting'.", + "input": "", + "output": "A synonym for 'exciting' is 'thrilling'." + }, + { + "instruction": "Suggest a more formal synonym for 'goodbye.'", + "input": "", + "output": "A more formal synonym for 'goodbye' is 'farewell.'" + }, + { + "instruction": "What is the pH of lemon juice?", + "input": "", + "output": "The pH of lemon juice is approximately 2.0." + }, + { + "instruction": "Rewrite the sentence in future perfect tense.", + "input": "They will complete the project.", + "output": "They will have completed the project." + }, + { + "instruction": "Re-word this sentence using an active voice.", + "input": "The decision was made by the team.", + "output": "The team made the decision." + }, + { + "instruction": "Provide an example of alliteration.", + "input": "", + "output": "An example of alliteration is, 'Peter Piper picked a peck of pickled peppers.'" + }, + { + "instruction": "What is the primary purpose of the United Nations?", + "input": "", + "output": "The primary purpose of the United Nations is to maintain international peace and security, develop friendly relations among nations, achieve international cooperation, and be a center for harmonizing the actions of nations." + }, + { + "instruction": "What is a synonym of 'astonishing'?", + "input": "", + "output": "A synonym for 'astonishing' is 'amazing'." + }, + { + "instruction": "What is the capital of Japan?", + "input": "", + "output": "The capital of Japan is Tokyo." + }, + { + "instruction": "Arrange the given numbers in descending order.", + "input": "7, 10, 3, 6, 9", + "output": "10, 9, 7, 6, 3." + }, + { + "instruction": "Name a punctuation mark used to indicate possession.", + "input": "", + "output": "A punctuation mark used to indicate possession is the apostrophe (')." + }, + { + "instruction": "Make the first sentence a dependent clause.", + "input": "He finished his homework. He went out to play.", + "output": "After he finished his homework, he went out to play." + }, + { + "instruction": "What is the opposite of 'cheap'?", + "input": "", + "output": "The opposite of 'cheap' is 'expensive'." + }, + { + "instruction": "What is the capital of Norway?", + "input": "", + "output": "The capital of Norway is Oslo." + }, + { + "instruction": "Categorize the following items into liquids and solids.", + "input": "Milk, Rock, Honey", + "output": "Liquids: Milk, Honey\nSolids: Rock" + }, + { + "instruction": "Translate 'Where are you from?' into French.", + "input": "", + "output": "The French translation of 'Where are you from?' is 'D'o\u00f9 viens-tu?'." + }, + { + "instruction": "Name a bird that cannot fly.", + "input": "", + "output": "A bird that cannot fly is the penguin." + }, + { + "instruction": "Generate a sentence using the word 'compassionate'.", + "input": "", + "output": "She is very compassionate and always helps those in need." + }, + { + "instruction": "What is the melting point of ice in Celsius?", + "input": "", + "output": "The melting point of ice is 0 degrees Celsius." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The night is very calm.", + "output": "The night is a tranquil sea." + }, + { + "instruction": "List the steps required to make a cup of tea.", + "input": "", + "output": "Steps to make a cup of tea: 1. Boil water. 2. Place a tea bag in a cup. 3. Pour hot water over the tea bag. 4. Let it steep. 5. Remove the tea bag and enjoy." + }, + { + "instruction": "Define the term 'osmosis'.", + "input": "", + "output": "Osmosis is the process of movement of water molecules through a semipermeable membrane from a region of low solute concentration to a region of high solute concentration." + }, + { + "instruction": "Create a compound word using 'water.'", + "input": "", + "output": "A compound word using 'water' is 'waterfall.'" + }, + { + "instruction": "Calculate the volume of a cube with side length of 3 meters.", + "input": "", + "output": "The volume of the cube is 27 cubic meters." + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just finished reading a fascinating book.", + "output": "What was the book about?" + }, + { + "instruction": "Generate a sentence using the word 'resilient'.", + "input": "", + "output": "He is very resilient and always bounces back from setbacks." + }, + { + "instruction": "Rewrite this sentence to start with 'In spite of': Although it was raining, they went for a walk.", + "input": "", + "output": "In spite of the rain, they went for a walk." + }, + { + "instruction": "Convert the mass from kilograms to grams.", + "input": "4 kilograms", + "output": "4 kilograms is equivalent to 4000 grams." + }, + { + "instruction": "Convert the number 110 from binary to decimal.", + "input": "", + "output": "The decimal equivalent of the binary number 110 is 6." + }, + { + "instruction": "Name three essential elements found in nature.", + "input": "", + "output": "1. Carbon\n2. Oxygen\n3. Iron" + }, + { + "instruction": "Provide a synonym for 'intelligent'.", + "input": "", + "output": "A synonym for 'intelligent' is 'smart'." + }, + { + "instruction": "What is the speed of sound?", + "input": "", + "output": "The speed of sound is approximately 343 meters per second at sea level and at 20\u00b0C." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very clumsy.", + "output": "He is as clumsy as a bull in a china shop." + }, + { + "instruction": "Find the temperature of boiling water under normal atmospheric pressure.", + "input": "", + "output": "Under normal atmospheric pressure, the temperature of boiling water is 100 degrees Celsius (212 degrees Fahrenheit)." + }, + { + "instruction": "Classify the following online review as being positive or negative.", + "input": "I found this book to be incredibly insightful and informative.", + "output": "The online review is positive." + }, + { + "instruction": "What is the pH level of pure water?", + "input": "", + "output": "The pH level of pure water is 7, which is neutral." + }, + { + "instruction": "What is the formula for calculating the perimeter of a rectangle?", + "input": "", + "output": "The formula for calculating the perimeter of a rectangle is 2(length + width)." + }, + { + "instruction": "Generate a sentence using the word 'determined'.", + "input": "", + "output": "She is determined to achieve her dreams." + }, + { + "instruction": "Generate a sentence using the word 'meticulous'.", + "input": "", + "output": "He is meticulous in his work and pays attention to every detail." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The room was very hot.", + "output": "The room was as hot as an oven." + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer is working perfectly.", + "output": "The statement 'My computer is working perfectly' can be classified as positive." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "We goes to the park every Sunday.", + "output": "We go to the park every Sunday." + }, + { + "instruction": "Define the term 'quantum mechanics'.", + "input": "", + "output": "Quantum mechanics is a branch of physics that studies the behavior of particles at the smallest scales, such as atoms and subatomic particles." + }, + { + "instruction": "What is the primary function of the digestive system?", + "input": "", + "output": "The primary function of the digestive system is to break down food into nutrients that can be absorbed into the bloodstream and used by the body for energy, growth, and repair." + }, + { + "instruction": "What is the formula for calculating the area of a rectangle?", + "input": "", + "output": "The formula for calculating the area of a rectangle is length \u00d7 width." + }, + { + "instruction": "Translate 'Thank you very much' into Japanese.", + "input": "", + "output": "The Japanese translation of 'Thank you very much' is '\u3069\u3046\u3082\u3042\u308a\u304c\u3068\u3046\u3054\u3056\u3044\u307e\u3059' (D\u014dmo arigat\u014dgozaimasu)." + }, + { + "instruction": "Translate the phrase 'How are you?' into Spanish.", + "input": "", + "output": "The Spanish translation of 'How are you?' is '\u00bfC\u00f3mo est\u00e1s?'" + }, + { + "instruction": "Fill in the blank to complete the sentence.", + "input": "The most popular vegetable in the world is the ___", + "output": "The most popular vegetable in the world is the potato." + }, + { + "instruction": "Given a sentence, count the number of words.", + "input": "I love watching the stars at night.", + "output": "The sentence 'I love watching the stars at night.' contains 7 words." + }, + { + "instruction": "Correct the grammatical errors in this sentence.", + "input": "He don't know nothing about it.", + "output": "He doesn't know anything about it." + }, + { + "instruction": "Suggest a more formal synonym for 'happy.'", + "input": "", + "output": "A more formal synonym for 'happy' is 'joyful.'" + }, + { + "instruction": "Convert 2,000 milligrams to grams.", + "input": "", + "output": "2,000 milligrams is equal to 2 grams." + }, + { + "instruction": "Put the following words in alphabetical order: kitten, zebra, alligator.", + "input": "", + "output": "Alligator, kitten, zebra." + }, + { + "instruction": "Identify the main theme in George Orwell's '1984'.", + "input": "", + "output": "The main theme of '1984' is the dangers of totalitarianism and oppressive government control." + }, + { + "instruction": "What is the capital of Germany?", + "input": "", + "output": "The capital city of Germany is Berlin." + }, + { + "instruction": "Define the term 'mitosis'.", + "input": "", + "output": "Mitosis is a type of cell division that results in two daughter cells each having the same number and kind of chromosomes as the parent nucleus." + }, + { + "instruction": "Name 3 different animals that are active during the day.", + "input": "", + "output": "1. Squirrel\n2. Eagle\n3. Tiger" + }, + { + "instruction": "Generate a sentence using the word 'spontaneous'.", + "input": "", + "output": "His decision to travel was completely spontaneous, made on a whim." + }, + { + "instruction": "Convert the sentence into a question by adding a question tag.", + "input": "You will help me", + "output": "You will help me, won't you?" + }, + { + "instruction": "What type of animal is a 'sparrow'?", + "input": "", + "output": "A sparrow is a type of bird." + }, + { + "instruction": "Create a compound word using 'sun.'", + "input": "", + "output": "A compound word using 'sun' is 'sunshine.'" + }, + { + "instruction": "Convert 90 centimeters to meters.", + "input": "", + "output": "90 centimeters is 0.9 meters." + }, + { + "instruction": "Provide a synonym for 'difficult'.", + "input": "", + "output": "A synonym for 'difficult' is 'challenging'." + }, + { + "instruction": "Classify the following into mammals, birds, and fish.", + "input": "Eagle, Shark, Lion", + "output": "Mammals: Lion\nBirds: Eagle\nFish: Shark" + }, + { + "instruction": "What is the square root of 16?", + "input": "", + "output": "The square root of 16 is 4." + }, + { + "instruction": "Translate 'We go to school together' into Spanish.", + "input": "", + "output": "Vamos juntos a la escuela." + }, + { + "instruction": "What is the melting point of tin in Celsius?", + "input": "", + "output": "The melting point of tin is 231.9 degrees Celsius." + }, + { + "instruction": "Edit the following sentence to make it more formal.", + "input": "I gotta go to the store.", + "output": "I have to go to the store." + }, + { + "instruction": "Translate 'Goodbye' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Goodbye' is 'Adi\u00f3s'." + }, + { + "instruction": "What is the chemical formula for nitric acid?", + "input": "", + "output": "The chemical formula for nitric acid is HNO3." + }, + { + "instruction": "Transform the following sentence to include an exclamation mark to show surprise.", + "input": "I found my keys", + "output": "I found my keys!" + }, + { + "instruction": "Generate a sentence using the word 'illuminate.'", + "input": "", + "output": "The streetlights illuminate the pathway at night." + }, + { + "instruction": "What is the chemical symbol for magnesium?", + "input": "", + "output": "The chemical symbol for magnesium is Mg." + }, + { + "instruction": "Convert the sentence to passive voice: 'The chef prepared the meal.'", + "input": "The chef prepared the meal.", + "output": "The meal was prepared by the chef." + }, + { + "instruction": "Edit the text for grammar and punctuation.", + "input": "today was amazing I went hiking and saw a bear", + "output": "\"Today was amazing. I went hiking and saw a bear.\"" + }, + { + "instruction": "Explain the function of the respiratory system in humans.", + "input": "", + "output": "The respiratory system's primary function is to supply the blood with oxygen through breathing and to remove carbon dioxide. It includes the nose, throat, larynx, trachea, bronchi, and lungs." + }, + { + "instruction": "Convert the number 10 from decimal to binary.", + "input": "", + "output": "The binary equivalent of the decimal number 10 is 1010." + }, + { + "instruction": "Rewrite the sentence so that it does not contain any passive voice.", + "input": "The song was played on the radio.", + "output": "The radio played the song." + }, + { + "instruction": "Generate a sentence using the word 'quintessential'.", + "input": "", + "output": "He was the quintessential example of a gentleman." + }, + { + "instruction": "What is the past tense of 'fly'?", + "input": "", + "output": "The past tense of 'fly' is 'flew'." + }, + { + "instruction": "Rewrite this statement as an imperative sentence.", + "input": "You should check the document for errors.", + "output": "Check the document for errors." + }, + { + "instruction": "Convert the following verb to its gerund form: 'write'", + "input": "", + "output": "The gerund form of 'write' is 'writing.'" + }, + { + "instruction": "Generate three verbs that mean the same as 'to write.'", + "input": "", + "output": "1. To compose\n2. To draft\n3. To pen" + }, + { + "instruction": "Calculate the perimeter of an equilateral triangle with each side measuring 8 cm.", + "input": "", + "output": "The perimeter of the equilateral triangle is 24 cm." + }, + { + "instruction": "Describe the color blue in 3 words.", + "input": "", + "output": "Deep, calming, cool." + }, + { + "instruction": "Classify the following items: apple, diamond, fern.", + "input": "", + "output": "Plants: Fern\nMinerals: Diamond\nFruits: Apple" + }, + { + "instruction": "Scramble the given phrase.", + "input": "A random phrase", + "output": "A nmaodr hrasep" + }, + { + "instruction": "Identify the verb in the following sentence", + "input": "The cat sleeps on the couch.", + "output": "The verb in the sentence is \"sleeps.\"" + }, + { + "instruction": "What is the freezing point of water in Celsius?", + "input": "", + "output": "The freezing point of water is 0 degrees Celsius." + }, + { + "instruction": "Pick the best synonym for the given word.", + "input": "Assist", + "output": "Help." + }, + { + "instruction": "Provide the past tense form of the verb 'choose.'", + "input": "", + "output": "The past tense of the verb 'choose' is 'chose.'" + }, + { + "instruction": "Convert 1000 millimeters to meters.", + "input": "", + "output": "1000 millimeters is 1 meter." + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just got a new job.", + "output": "What will you be doing at your new job?" + }, + { + "instruction": "Given a sentence, convert it into indirect speech.", + "input": "He said, 'I am hungry.'", + "output": "He said that he was hungry." + }, + { + "instruction": "What is the integral of x?", + "input": "", + "output": "The integral of x is (1/2)x^2 + C." + }, + { + "instruction": "Name the process by which leaves change color in autumn.", + "input": "", + "output": "The process by which leaves change color in autumn is called senescence." + }, + { + "instruction": "What is the capital of Mexico?", + "input": "", + "output": "The capital of Mexico is Mexico City." + }, + { + "instruction": "Categorize the following items into one of the following categories: appliance, clothing, or food.", + "input": "Microwave, T-shirt, Bread", + "output": "Appliance: Microwave\nClothing: T-shirt\nFood: Bread" + }, + { + "instruction": "Translate 'I am sorry' into German.", + "input": "", + "output": "The German translation of 'I am sorry' is 'Es tut mir leid'." + }, + { + "instruction": "Convert this declarative sentence into a command: 'You should check the report.'", + "input": "", + "output": "Check the report." + }, + { + "instruction": "Translate 'good afternoon' into Portuguese.", + "input": "", + "output": "Boa tarde" + }, + { + "instruction": "Convert the following verb to its gerund form: 'eat'", + "input": "", + "output": "The gerund form of 'eat' is 'eating'." + }, + { + "instruction": "Create a sentence using the word 'bittersweet'.", + "input": "", + "output": "His victory was bittersweet, as it came at a great personal loss." + }, + { + "instruction": "What is the opposite of 'strong'?", + "input": "", + "output": "The opposite of 'strong' is 'weak'." + }, + { + "instruction": "Translate 'They are reading books' into Spanish.", + "input": "", + "output": "Est\u00e1n leyendo libros." + }, + { + "instruction": "List two synonyms for 'beautiful.'", + "input": "", + "output": "Two synonyms for 'beautiful' are 'gorgeous' and 'stunning.'" + }, + { + "instruction": "Rewrite this sentence to increase formality: 'Kids need to check their phones less.'", + "input": "", + "output": "Children ought to reduce their mobile phone usage." + }, + { + "instruction": "What is the opposite of 'slow'?", + "input": "", + "output": "The opposite of 'slow' is 'fast'." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The baby is very cute.", + "output": "The baby is as cute as a button." + }, + { + "instruction": "Determine the state of matter for helium at room temperature.", + "input": "", + "output": "Helium is in a gaseous state at room temperature." + }, + { + "instruction": "Correct the misspelled word in the following sentence.", + "input": "I recieved your message", + "output": "The misspelled word \"recieved\" should be spelled \"received.\"" + }, + { + "instruction": "What is the formula for kinetic energy?", + "input": "", + "output": "The formula for kinetic energy is KE = 1/2 mv\u00b2, where m is mass and v is velocity." + }, + { + "instruction": "Identify the adjective in the sentence.", + "input": "The red car sped down the road.", + "output": "The adjective in the sentence is 'red'." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'Life is like a journey.'", + "input": "", + "output": "Life is a journey." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very gentle.", + "output": "She is as gentle as a lamb." + }, + { + "instruction": "Change the given verb to its present perfect form.", + "input": "go", + "output": "Gone." + }, + { + "instruction": "Convert 200 grams to kilograms.", + "input": "", + "output": "200 grams is 0.2 kilograms." + }, + { + "instruction": "Determine the area of a square with side length of 8 meters.", + "input": "", + "output": "The area of the square is 64 square meters." + }, + { + "instruction": "Reverse this array", + "input": "[5, 15, 25, 35, 45]", + "output": "Here is the reversed array: [45, 35, 25, 15, 5]" + }, + { + "instruction": "Generate a random sequence of 12 digits.", + "input": "", + "output": "Here is a randomly generated sequence of 12 digits: 728493651280." + }, + { + "instruction": "Create a simile with the phrase 'as busy as'.", + "input": "", + "output": "He was as busy as a bee." + }, + { + "instruction": "What is the opposite of 'noisy'?", + "input": "", + "output": "The opposite of 'noisy' is 'quiet'." + }, + { + "instruction": "Generate a random alphanumeric code of 8 characters.", + "input": "", + "output": "Here is a randomly generated 8-character alphanumeric code: A3F8J7K2." + }, + { + "instruction": "Name three essential minerals for human health.", + "input": "", + "output": "1. Calcium\n2. Iron\n3. Magnesium" + }, + { + "instruction": "Categorize the following items into herbs and spices.", + "input": "Cinnamon, Basil, Pepper", + "output": "Herbs: Basil\nSpices: Cinnamon, Pepper" + }, + { + "instruction": "Create a sentence that begins with 'Even though,'", + "input": "", + "output": "Even though it was raining, they decided to go for a walk." + }, + { + "instruction": "What is the opposite of 'happy'?", + "input": "", + "output": "The opposite of 'happy' is 'sad'." + }, + { + "instruction": "Define the term 'allusion'.", + "input": "", + "output": "An allusion is a brief and indirect reference to a person, place, thing, or idea of historical, cultural, literary, or political significance." + }, + { + "instruction": "Classify the item as either a mammal or a bird.", + "input": "Penguin", + "output": "Bird" + }, + { + "instruction": "Convert the temperature from Celsius to Fahrenheit.", + "input": "30 degrees Celsius", + "output": "30 degrees Celsius is 86 degrees Fahrenheit." + }, + { + "instruction": "Convert 15 liters to milliliters.", + "input": "", + "output": "15 liters is 15000 milliliters." + }, + { + "instruction": "Combine the given sentences into one cohesive sentence.", + "input": "It is raining. I cannot go to the park.", + "output": "Since it is raining, I cannot go to the park." + }, + { + "instruction": "Calculate the future value of an investment of $500 at an annual interest rate of 5% compounded yearly for 10 years.", + "input": "", + "output": "The future value of the investment would be approximately $814.45." + }, + { + "instruction": "What is a synonym of 'lazy'?", + "input": "", + "output": "A synonym for 'lazy' is 'indolent'." + }, + { + "instruction": "Look up the boiling point of ethanol.", + "input": "", + "output": "The boiling point of ethanol is approximately 78.37 degrees Celsius or 173.07 degrees Fahrenheit." + }, + { + "instruction": "Convert the sentence to passive voice: 'The teacher explains the lesson.'", + "input": "The teacher explains the lesson.", + "output": "The lesson is explained by the teacher." + }, + { + "instruction": "Name the largest ocean on Earth.", + "input": "", + "output": "The largest ocean on Earth is the Pacific Ocean." + }, + { + "instruction": "What is the boiling point of water on Mount Everest?", + "input": "", + "output": "The boiling point of water on Mount Everest is lower than at sea level, around 71 degrees Celsius." + }, + { + "instruction": "Evaluate the following phrase by transforming it into the correct spelling given.", + "input": "recieve --> receive", + "output": "The spelling of the given phrase 'recieve' is incorrect, the correct spelling is 'receive.'" + }, + { + "instruction": "Define the term 'metaphor'.", + "input": "", + "output": "A metaphor is a figure of speech in which a word or phrase is applied to an object or action to which it is not literally applicable." + }, + { + "instruction": "Explain what GDP stands for and what it measures.", + "input": "", + "output": "GDP stands for Gross Domestic Product, which measures the total value of all goods and services produced over a specific time period within a country's borders." + }, + { + "instruction": "Provide a synonym for 'delicious'.", + "input": "", + "output": "A synonym for 'delicious' is 'tasty'." + }, + { + "instruction": "Translate this phrase to Spanish: 'I will travel tomorrow.'", + "input": "", + "output": "Viajar\u00e9 ma\u00f1ana." + }, + { + "instruction": "What is a synonym for 'enormous'?", + "input": "", + "output": "A synonym for 'enormous' is 'gigantic'." + }, + { + "instruction": "Create a new sentence by combining the two sentences.", + "input": "He likes to swim. He goes to the pool every day.", + "output": "He likes to swim and goes to the pool every day." + }, + { + "instruction": "What is the square root of 169?", + "input": "", + "output": "The square root of 169 is 13." + }, + { + "instruction": "Identify the genre of the book '1984' by George Orwell.", + "input": "", + "output": "The genre of '1984' by George Orwell is dystopian fiction." + }, + { + "instruction": "Convert 120 meters to kilometers.", + "input": "", + "output": "120 meters is 0.12 kilometers." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "The end result was a total success.", + "output": "The result was a success." + }, + { + "instruction": "Name a punctuation mark used to indicate a pause.", + "input": "", + "output": "A punctuation mark used to indicate a pause is a comma (,)." + }, + { + "instruction": "Suggest a simpler word to replace the word in braces.", + "input": "He moved {expeditiously} through the crowd.", + "output": "He moved quickly through the crowd." + }, + { + "instruction": "Construct a sentence using the words 'achieve,' 'goal,' and 'dedication.'", + "input": "", + "output": "With dedication, you can achieve any goal." + }, + { + "instruction": "Generate a question based on the following statement.", + "input": "Plants produce oxygen through the process of photosynthesis.", + "output": "How do plants produce oxygen?" + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The night is very dark.", + "output": "The night is a black canvas." + }, + { + "instruction": "Translate 'good evening' into Italian.", + "input": "", + "output": "The Italian translation of 'good evening' is 'buona sera'." + }, + { + "instruction": "What is the cube root of 27?", + "input": "", + "output": "The cube root of 27 is 3." + }, + { + "instruction": "What is the atomic number of nitrogen?", + "input": "", + "output": "The atomic number of nitrogen is 7." + }, + { + "instruction": "Name 3 different animals that are nocturnal.", + "input": "", + "output": "1. Owl\n2. Bat\n3. Raccoon" + }, + { + "instruction": "Convert 100 grams to kilograms.", + "input": "", + "output": "100 grams is 0.1 kilograms." + }, + { + "instruction": "Provide a synonym for 'brave'.", + "input": "", + "output": "A synonym for 'brave' is 'fearless'." + }, + { + "instruction": "Convert 12 meters to centimeters.", + "input": "", + "output": "12 meters is 1200 centimeters." + }, + { + "instruction": "What is the pH of a substance that is neither acidic nor basic?", + "input": "", + "output": "The pH of a substance that is neither acidic nor basic is 7, which is neutral." + }, + { + "instruction": "Create a new sentence by combining the two sentences.", + "input": "She loves to read. She goes to the library every week.", + "output": "She loves to read and goes to the library every week." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "He is very busy.", + "output": "He is a bee in a hive." + }, + { + "instruction": "What is a synonym of 'exquisite'?", + "input": "", + "output": "A synonym for 'exquisite' is 'elegant'." + }, + { + "instruction": "Convert the following sentence to use a conditional clause: If it rains, we will stay indoors.", + "input": "", + "output": "We will stay indoors if it rains." + }, + { + "instruction": "What is the melting point of mercury in Celsius?", + "input": "", + "output": "The melting point of mercury is -38.83 degrees Celsius." + }, + { + "instruction": "Generate a sentence using the word 'clarify.'", + "input": "", + "output": "She asked the teacher to clarify the instructions." + }, + { + "instruction": "Generate a sentence using the word 'meticulous'.", + "input": "", + "output": "He was meticulous in his planning for the project." + }, + { + "instruction": "What is the capital of Italy?", + "input": "", + "output": "The capital of Italy is Rome." + }, + { + "instruction": "Rewrite the following sentence using the active voice.", + "input": "The ball was thrown by the pitcher.", + "output": "The pitcher threw the ball." + }, + { + "instruction": "What is the melting point of copper in Celsius?", + "input": "", + "output": "The melting point of copper is 1085 degrees Celsius." + }, + { + "instruction": "What is the capital of the Netherlands?", + "input": "", + "output": "The capital of the Netherlands is Amsterdam." + }, + { + "instruction": "What is the chemical symbol for sodium?", + "input": "", + "output": "The chemical symbol for sodium is Na." + }, + { + "instruction": "Rewrite the following sentence by adding one adjective.", + "input": "The cat slept on the mat.", + "output": "The black cat slept on the mat." + }, + { + "instruction": "Explain what an enzyme is.", + "input": "", + "output": "An enzyme is a protein that catalyzes chemical reactions within the body." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The night is very peaceful.", + "output": "The night is a blanket of calm." + }, + { + "instruction": "Identify the protagonist in 'The Great Gatsby'.", + "input": "", + "output": "The protagonist of 'The Great Gatsby' is Jay Gatsby." + }, + { + "instruction": "Categorize the following items into one of the following categories: tool, clothing, or fruit.", + "input": "Hammer, Shirt, Apple", + "output": "Tool: Hammer\nClothing: Shirt\nFruit: Apple" + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very quiet.", + "output": "She is as quiet as a mouse." + }, + { + "instruction": "What type of molecule is DNA?", + "input": "", + "output": "DNA is a type of nucleic acid." + }, + { + "instruction": "Provide the plural form of 'index.'", + "input": "", + "output": "The plural form of 'index' is 'indices.'" + }, + { + "instruction": "What is the unit for measuring force?", + "input": "", + "output": "The unit for measuring force is the newton." + }, + { + "instruction": "Give me a synonym for the word 'strong'.", + "input": "", + "output": "Powerful." + }, + { + "instruction": "What is the plural form of 'bacterium'?", + "input": "", + "output": "The plural form of 'bacterium' is 'bacteria'." + }, + { + "instruction": "What is an antonym of 'expand'?", + "input": "", + "output": "An antonym of 'expand' is 'contract'." + }, + { + "instruction": "What is the primary function of the human nervous system?", + "input": "", + "output": "The primary function of the human nervous system is to control and coordinate the body's activities by transmitting signals between different parts of the body. It includes the brain, spinal cord, and peripheral nerves." + }, + { + "instruction": "What is the boiling point of nitrogen?", + "input": "", + "output": "The boiling point of nitrogen is -195.79 degrees Celsius." + }, + { + "instruction": "Translate the term 'global warming' into Spanish.", + "input": "", + "output": "The Spanish translation for 'global warming' is 'calentamiento global'." + }, + { + "instruction": "Convert 20 miles to kilometers, assuming 1 mile is approximately 1.60934 kilometers.", + "input": "", + "output": "20 miles is approximately 32.1868 kilometers." + }, + { + "instruction": "What type of animal is a 'penguin'?", + "input": "", + "output": "A penguin is a type of bird." + }, + { + "instruction": "What is the opposite of 'exclude'?", + "input": "", + "output": "The opposite of 'exclude' is 'include'." + }, + { + "instruction": "List three primary colors.", + "input": "", + "output": "The three primary colors are red, blue, and yellow." + }, + { + "instruction": "Create a sentence that begins with \"If only,\"", + "input": "", + "output": "If only I had known the truth earlier..." + }, + { + "instruction": "What is the opposite of 'clean'?", + "input": "", + "output": "The opposite of 'clean' is 'dirty'." + }, + { + "instruction": "Rewrite the given passage in the past tense.", + "input": "They are building a new library.", + "output": "They built a new library." + }, + { + "instruction": "Name an antonym for \"joy.\"", + "input": "", + "output": "An antonym for \"joy\" is \"sorrow.\"" + }, + { + "instruction": "Provide an example of a complex sentence.", + "input": "", + "output": "Although it was raining, we decided to go for a walk." + }, + { + "instruction": "Name the three main types of rocks.", + "input": "", + "output": "The three main types of rocks are igneous, sedimentary, and metamorphic." + }, + { + "instruction": "What is the boiling point of ethanol in Celsius?", + "input": "", + "output": "The boiling point of ethanol is 78.37 degrees Celsius." + }, + { + "instruction": "Convert 45 kilograms to pounds.", + "input": "", + "output": "45 kilograms is approximately 99.21 pounds." + }, + { + "instruction": "Classify this sentence into one of the following topics: health, economy, technology, music", + "input": "Electric cars are revolutionizing the automotive industry.", + "output": "The topic of the given sentence is technology." + }, + { + "instruction": "What is the largest planet in our solar system?", + "input": "", + "output": "The largest planet in our solar system is Jupiter." + }, + { + "instruction": "What is the capital of Australia?", + "input": "", + "output": "The capital of Australia is Canberra." + }, + { + "instruction": "Determine whether the given two words are synonyms or antonyms.", + "input": "Start - Stop", + "output": "Start and Stop are antonyms." + }, + { + "instruction": "Name the literary device used in 'Time is a thief.'", + "input": "", + "output": "The literary device used is metaphor." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: The sun set behind the hills.", + "input": "", + "output": "The sun dipped behind the hills like a shy child hiding." + }, + { + "instruction": "List the primary colors in the CMYK color model.", + "input": "", + "output": "The primary colors in the CMYK color model are cyan, magenta, yellow, and key (black)." + }, + { + "instruction": "Classify the following animal as carnivore, herbivore, or omnivore.", + "input": "Bear", + "output": "Omnivore" + }, + { + "instruction": "List 5 different types of fruits", + "input": "", + "output": "1. Apple\n2. Banana\n3. Orange\n4. Strawberry\n5. Mango" + }, + { + "instruction": "What is the sum of 2 and 3?", + "input": "", + "output": "The sum of 2 and 3 is 5." + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Banana, Apple, Cherry", + "output": "sorted in alphabetical order: Apple, Banana, Cherry" + }, + { + "instruction": "What is the boiling point of water in Fahrenheit?", + "input": "", + "output": "The boiling point of water is 212 degrees Fahrenheit." + }, + { + "instruction": "Change this sentence into the present perfect tense.", + "input": "She writes a letter.", + "output": "She has written a letter." + }, + { + "instruction": "Convert 100 centimeters to meters.", + "input": "", + "output": "100 centimeters is equal to 1 meter." + }, + { + "instruction": "Identify the hypernym for the word \"car\".", + "input": "", + "output": "The hypernym for the word \"car\" is \"vehicle\"." + }, + { + "instruction": "Generate a sentence using the word 'dynamic'.", + "input": "", + "output": "The team had a dynamic leader who inspired everyone." + }, + { + "instruction": "Rewrite this English sentence into passive voice.", + "input": "The gardener planted the flowers.", + "output": "The flowers were planted by the gardener." + }, + { + "instruction": "Convert the active sentence to passive: 'We celebrate Independence Day on July 4th.'", + "input": "", + "output": "Independence Day is celebrated on July 4th." + }, + { + "instruction": "Convert 800 grams to kilograms.", + "input": "", + "output": "800 grams is 0.8 kilograms." + }, + { + "instruction": "Convert 15 feet to meters.", + "input": "", + "output": "15 feet is equal to approximately 4.572 meters." + }, + { + "instruction": "Translate 'How are you?' into Japanese.", + "input": "", + "output": "The Japanese translation of 'How are you?' is '\u304a\u5143\u6c17\u3067\u3059\u304b' (Ogenki desu ka)." + }, + { + "instruction": "Generate a sentence using the word 'enthusiastic'.", + "input": "", + "output": "He was enthusiastic about starting his new job." + }, + { + "instruction": "What is the opposite of 'cold'?", + "input": "", + "output": "The opposite of 'cold' is 'hot'." + }, + { + "instruction": "What is the boiling point of propane in Celsius?", + "input": "", + "output": "The boiling point of propane is -42 degrees Celsius." + }, + { + "instruction": "What is the square root of 81?", + "input": "", + "output": "The square root of 81 is 9." + }, + { + "instruction": "Generate a sentence using the word 'effort'.", + "input": "", + "output": "She put a lot of effort into her work." + }, + { + "instruction": "Provide a synonym for 'swift'.", + "input": "", + "output": "A synonym for 'swift' is 'quick'." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Ice, Coffee, Helium", + "output": "Ice - Solid\nCoffee - Liquid\nHelium - Gas" + }, + { + "instruction": "Translate 'What time is it?' into Italian.", + "input": "", + "output": "The Italian translation of 'What time is it?' is 'Che ora \u00e8?'." + }, + { + "instruction": "Transform the following sentence into a question using 'would.'", + "input": "You can lend me a pen.", + "output": "Would you lend me a pen?" + }, + { + "instruction": "Describe the process of condensation.", + "input": "", + "output": "Condensation is the process by which water vapor in the air is changed into liquid water." + }, + { + "instruction": "Correct any spelling mistakes in the given sentence.", + "input": "The beutiful garden was filled with flowers.", + "output": "The beautiful garden was filled with flowers." + }, + { + "instruction": "Calculate the perimeter of a hexagon with side lengths of 5 cm.", + "input": "", + "output": "The perimeter of the hexagon is 30 cm." + }, + { + "instruction": "Convert 30 kilometers to miles.", + "input": "", + "output": "30 kilometers is approximately 18.64 miles." + }, + { + "instruction": "Define the term 'algorithm'.", + "input": "", + "output": "An algorithm is a set of defined, step-by-step procedures or rules to solve a problem or accomplish a task." + }, + { + "instruction": "Convert 500 grams to ounces.", + "input": "", + "output": "500 grams is approximately 17.64 ounces." + }, + { + "instruction": "Convert 500 grams to kilograms.", + "input": "", + "output": "500 grams is 0.5 kilograms." + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Necesary", + "output": "The correct spelling is \"Necessary\"." + }, + { + "instruction": "What is the opposite of 'light'?", + "input": "", + "output": "The opposite of 'light' is 'heavy'." + }, + { + "instruction": "What is the capital of Germany?", + "input": "", + "output": "The capital of Germany is Berlin." + }, + { + "instruction": "Explain what a black hole is.", + "input": "", + "output": "A black hole is a region of space where gravity is so strong that nothing, not even light, can escape from it." + }, + { + "instruction": "What is the symbol for copper on the periodic table?", + "input": "", + "output": "The symbol for copper on the periodic table is Cu." + }, + { + "instruction": "Fill in the blanks with appropriate words.", + "input": "The sky was filled with colorful ________ at sunrise.", + "output": "The sky was filled with colorful hues at sunrise." + }, + { + "instruction": "Convert the number 100 from binary to decimal.", + "input": "", + "output": "The decimal equivalent of the binary number 100 is 4." + }, + { + "instruction": "What is the chemical formula for hydrochloric acid?", + "input": "", + "output": "The chemical formula for hydrochloric acid is HCl." + }, + { + "instruction": "Rewrite the sentence.", + "input": "The poem was composed in a lyrical manner.", + "output": "The poem was crafted lyrically." + }, + { + "instruction": "What is the square root of 144?", + "input": "", + "output": "The square root of 144 is 12." + }, + { + "instruction": "What is the formula for converting Celsius to Fahrenheit?", + "input": "", + "output": "The formula for converting Celsius to Fahrenheit is (C \u00d7 9/5) + 32." + }, + { + "instruction": "What is the chemical formula for sodium carbonate?", + "input": "", + "output": "The chemical formula for sodium carbonate is Na2CO3." + }, + { + "instruction": "Convert 9 kilometers to meters.", + "input": "", + "output": "9 kilometers is 9000 meters." + }, + { + "instruction": "Convert the following verb to its infinitive form", + "input": "went", + "output": "The infinitive form of 'went' is 'to go'." + }, + { + "instruction": "Generate a sentence using the word 'diligent'.", + "input": "", + "output": "He is very diligent in his studies." + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer crashed.", + "output": "Negative" + }, + { + "instruction": "What is the pH of vinegar?", + "input": "", + "output": "The pH of vinegar is typically around 2.5." + }, + { + "instruction": "Calculate the area of a triangle with base 10 meters and height 5 meters.", + "input": "", + "output": "The area of the triangle is 25 square meters." + }, + { + "instruction": "Identify the primary colors in the RYB color model.", + "input": "", + "output": "The primary colors in the RYB color model are red, yellow, and blue." + }, + { + "instruction": "What is the capital of Spain?", + "input": "", + "output": "The capital of Spain is Madrid." + }, + { + "instruction": "Identify the adjective in the sentence.", + "input": "The large dog barked loudly.", + "output": "The adjective in the sentence is 'large'." + }, + { + "instruction": "Translate 'She studies chemistry' into German.", + "input": "", + "output": "Sie studiert Chemie." + }, + { + "instruction": "What is the comparative form of 'bad'?", + "input": "", + "output": "The comparative form of 'bad' is 'worse.'" + }, + { + "instruction": "Generate a sentence using the word 'charismatic'.", + "input": "", + "output": "The politician was very charismatic and attracted many followers." + }, + { + "instruction": "Convert 10 meters to centimeters.", + "input": "", + "output": "10 meters is 1000 centimeters." + }, + { + "instruction": "Generate three adjectives to describe a desert.", + "input": "", + "output": "1. Arid\n2. Vast\n3. Barren" + }, + { + "instruction": "Provide a synonym for 'quick'.", + "input": "", + "output": "A synonym for 'quick' is 'fast'." + }, + { + "instruction": "Find a synonym for the word 'begin.'", + "input": "", + "output": "A synonym for the word 'begin' could be 'start.'" + }, + { + "instruction": "Define the term 'democracy'.", + "input": "", + "output": "Democracy is a system of government where the citizens exercise power by voting. It typically features fair and free elections, protection of human rights, and the rule of law." + }, + { + "instruction": "What is the opposite of 'expensive'?", + "input": "", + "output": "The opposite of 'expensive' is 'cheap'." + }, + { + "instruction": "Identify the interjection in the sentence.", + "input": "Ouch, that hurt!", + "output": "The interjection in the sentence is 'Ouch.'" + }, + { + "instruction": "Define the term 'photosynthesis'.", + "input": "", + "output": "Photosynthesis is the process by which green plants use sunlight to synthesize foods from carbon dioxide and water." + }, + { + "instruction": "Rewrite this English sentence into passive voice.", + "input": "Alex fixed the car.", + "output": "The car was fixed by Alex." + }, + { + "instruction": "Determine the product of 9 and 6.", + "input": "", + "output": "The product of 9 and 6 is 54." + }, + { + "instruction": "Calculate the compound interest on a principal amount of $1000 at an annual rate of 4% compounded annually for 5 years.", + "input": "", + "output": "The compound interest on $1000 at 4% per annum compounded annually for 5 years is approximately $216.65, making the total $1216.65." + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just returned from a vacation in Spain.", + "output": "How was your vacation in Spain?" + }, + { + "instruction": "What is the primary purpose of photosynthesis?", + "input": "", + "output": "The primary purpose of photosynthesis is to convert light energy into chemical energy stored in glucose, which can be used by plants to fuel their activities." + }, + { + "instruction": "What is the opposite of 'gain'?", + "input": "", + "output": "The opposite of 'gain' is 'lose.'" + }, + { + "instruction": "Convert 25 meters to centimeters.", + "input": "", + "output": "25 meters is 2500 centimeters." + }, + { + "instruction": "What is the chemical formula for glucose?", + "input": "", + "output": "The chemical formula for glucose is C6H12O6." + }, + { + "instruction": "What is a synonym of 'happy'?", + "input": "", + "output": "A synonym for 'happy' is 'joyful'." + }, + { + "instruction": "What type of animal is a 'whale'?", + "input": "", + "output": "A whale is a type of mammal." + }, + { + "instruction": "Translate 'How much does this cost?' into German.", + "input": "", + "output": "The German translation of 'How much does this cost?' is 'Wie viel kostet das?'." + }, + { + "instruction": "What is the chemical formula for methane?", + "input": "", + "output": "The chemical formula for methane is CH4." + }, + { + "instruction": "What is the derivative of x^2?", + "input": "", + "output": "The derivative of x^2 is 2x." + }, + { + "instruction": "What is the chemical symbol for iron?", + "input": "", + "output": "The chemical symbol for iron is Fe." + }, + { + "instruction": "Identify the setting in 'The Great Gatsby'.", + "input": "", + "output": "The setting of 'The Great Gatsby' is in the fictional towns of West Egg and East Egg on Long Island in the summer of 1922." + }, + { + "instruction": "Rewrite the following sentence so that it is in passive voice.", + "input": "The children played the game.", + "output": "The game was played by the children." + }, + { + "instruction": "Convert 1 liter to milliliters.", + "input": "", + "output": "1 liter is 1000 milliliters." + }, + { + "instruction": "Identify the main verb in the sentence.", + "input": "The cat chased the mouse.", + "output": "The main verb in the sentence is 'chased.'" + }, + { + "instruction": "What is the opposite of 'ascend'?", + "input": "", + "output": "The opposite of 'ascend' is 'descend'." + }, + { + "instruction": "Translate 'Thank you for your help' into Mandarin Chinese.", + "input": "", + "output": "The Mandarin Chinese translation of 'Thank you for your help' is '\u8c22\u8c22\u4f60\u7684\u5e2e\u52a9' (Xi\u00e8xi\u00e8 n\u01d0 de b\u0101ngzh\u00f9)." + }, + { + "instruction": "Name the composer of the Brandenburg Concertos.", + "input": "", + "output": "The composer of the Brandenburg Concertos is Johann Sebastian Bach." + }, + { + "instruction": "What is a synonym for 'assist'?", + "input": "", + "output": "A synonym for 'assist' is 'help'." + }, + { + "instruction": "Identify the type of sentence.", + "input": "Close the door.", + "output": "The type of sentence is imperative." + }, + { + "instruction": "Classify the following items: laptop, sunflower, tiger.", + "input": "", + "output": "Animals: Tiger\nPlants: Sunflower\nElectronics: Laptop" + }, + { + "instruction": "Name the process by which water changes from solid to liquid.", + "input": "", + "output": "The process by which water changes from solid to liquid is called melting." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'stunning'." + }, + { + "instruction": "Generate three adjectives to describe a mountain.", + "input": "", + "output": "1. Majestic\n2. Rugged\n3. Towering" + }, + { + "instruction": "Translate 'Can you help me?' into Spanish.", + "input": "", + "output": "\u00bfPuedes ayudarme?" + }, + { + "instruction": "Find a rhyming word for the word 'house.'", + "input": "", + "output": "A rhyming word for the word 'house' is 'mouse.'" + }, + { + "instruction": "Convert 3 liters to milliliters.", + "input": "", + "output": "3 liters is 3000 milliliters." + }, + { + "instruction": "Alter the content of the sentence to use the past tense.", + "input": "The train departs at noon.", + "output": "The train departed at noon." + }, + { + "instruction": "Generate a sentence using the word 'brave'.", + "input": "", + "output": "The firefighter was very brave and rescued the child from the fire." + }, + { + "instruction": "Translate 'How are you?' into French.", + "input": "", + "output": "The French translation of 'How are you?' is 'Comment \u00e7a va?'." + }, + { + "instruction": "Rearrange the following list into descending order.", + "input": "5, -3, 20, 10", + "output": "20, 10, 5, -3" + }, + { + "instruction": "Generate a sentence using the word 'perseverant'.", + "input": "", + "output": "He is perseverant and never gives up on his goals." + }, + { + "instruction": "Correct the punctuation in the sentence.", + "input": "Lets eat, Grandma!", + "output": "The corrected sentence should be: 'Let's eat, Grandma!'" + }, + { + "instruction": "Fix the following grammar mistake.", + "input": "They wants to go to the movies, but they have no money.", + "output": "They want to go to the movies, but they have no money." + }, + { + "instruction": "What is the pH of soap?", + "input": "", + "output": "The pH of soap is typically between 9 and 10." + }, + { + "instruction": "Classify the following items: pine tree, quartz, salmon.", + "input": "", + "output": "Plants: Pine tree\nMinerals: Quartz\nAnimals: Salmon" + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The meal was not delicious.", + "output": "The meal could use some improvement." + }, + { + "instruction": "What is the opposite of 'full'?", + "input": "", + "output": "The opposite of 'full' is 'empty'." + }, + { + "instruction": "Generate an antonym for the word 'calm.'", + "input": "", + "output": "An antonym for the word 'calm' could be 'agitated.'" + }, + { + "instruction": "What is the chemical formula for calcium carbonate?", + "input": "", + "output": "The chemical formula for calcium carbonate is CaCO3." + }, + { + "instruction": "Create a compound word using \"book.\"", + "input": "", + "output": "A compound word using \"book\" is \"bookshelf.\"" + }, + { + "instruction": "What is the chemical formula for hydrogen peroxide?", + "input": "", + "output": "The chemical formula for hydrogen peroxide is H2O2." + }, + { + "instruction": "What is the abbreviation for 'Bachelor of Science'?", + "input": "", + "output": "The abbreviation for 'Bachelor of Science' is B.Sc." + }, + { + "instruction": "Convert 30 centimeters to meters.", + "input": "", + "output": "30 centimeters is 0.3 meters." + }, + { + "instruction": "What is the boiling point of helium in Celsius?", + "input": "", + "output": "The boiling point of helium is -268.9 degrees Celsius." + }, + { + "instruction": "Convert 2 liters to milliliters.", + "input": "", + "output": "2 liters is 2000 milliliters." + }, + { + "instruction": "Transform the following sentence to include an exclamation mark to show excitement.", + "input": "I won the lottery", + "output": "I won the lottery!" + }, + { + "instruction": "Rewrite the following sentence to use a simile: The wind howled through the night.", + "input": "", + "output": "The wind howled through the night like a pack of wolves." + }, + { + "instruction": "What is the pH value of pure water at 25\u00b0C?", + "input": "", + "output": "The pH value of pure water at 25\u00b0C is 7." + }, + { + "instruction": "Arrange these historical periods in chronological order: Middle Ages, Renaissance, Industrial Revolution.", + "input": "", + "output": "1. Middle Ages\n2. Renaissance\n3. Industrial Revolution" + }, + { + "instruction": "Find a synonym for the word 'start'.", + "input": "", + "output": "A synonym for the word 'start' could be \"commence.\"" + }, + { + "instruction": "Identify the primary colors in the RGB color model.", + "input": "", + "output": "The primary colors in the RGB color model are red, green, and blue." + }, + { + "instruction": "Name three types of clouds.", + "input": "", + "output": "Three types of clouds are cumulus, stratus, and cirrus." + }, + { + "instruction": "Identify the tone used in the poem 'The Road Not Taken' by Robert Frost.", + "input": "", + "output": "The tone of 'The Road Not Taken' is reflective and contemplative, as the speaker considers the choices made in life and their long-term impacts." + }, + { + "instruction": "Change this sentence into the past perfect continuous tense.", + "input": "She sings in the choir.", + "output": "She had been singing in the choir." + }, + { + "instruction": "Generate a sentence using the word 'persistent'.", + "input": "", + "output": "He is very persistent and never gives up." + }, + { + "instruction": "Generate a sentence using the word 'curious'.", + "input": "", + "output": "The child was curious about everything around him." + }, + { + "instruction": "What is the boiling point of bromine in Celsius?", + "input": "", + "output": "The boiling point of bromine is 59 degrees Celsius." + }, + { + "instruction": "Convert the phrase into a complete sentence.", + "input": "sunset over the ocean", + "output": "The sunset over the ocean was breathtaking." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The room was very small.", + "output": "The room was a shoebox." + }, + { + "instruction": "Arrange the following events in chronological order: First Moon Landing, French Revolution, Industrial Revolution.", + "input": "", + "output": "1. French Revolution\n2. Industrial Revolution\n3. First Moon Landing" + }, + { + "instruction": "What is the capital of the United States?", + "input": "", + "output": "The capital of the United States is Washington, D.C." + }, + { + "instruction": "Describe the main function of the human skeletal system.", + "input": "", + "output": "The main function of the human skeletal system is to provide support and structure to the body, protect internal organs, enable movement through attachment points for muscles, and produce blood cells in the bone marrow." + }, + { + "instruction": "Identify the tense used in the sentence.", + "input": "They are watching a movie.", + "output": "The tense used in the sentence \"They are watching a movie\" is the present continuous tense." + }, + { + "instruction": "Translate 'thank you' into German.", + "input": "", + "output": "The German translation of 'thank you' is 'danke'." + }, + { + "instruction": "Suggest a continuation for the sentence.", + "input": "After the rain stopped,", + "output": "'the children went outside to play.'" + }, + { + "instruction": "What is the freezing point of water in Fahrenheit?", + "input": "", + "output": "The freezing point of water is 32 degrees Fahrenheit." + }, + { + "instruction": "Generate a sentence that follows the pattern: 'Under no circumstances should you _____'", + "input": "", + "output": "Under no circumstances should you open the door." + }, + { + "instruction": "Fix the following grammar mistake.", + "input": "He want to go to the movies, but he has no money.", + "output": "He wants to go to the movies, but he has no money." + }, + { + "instruction": "What is a synonym for 'cold'?", + "input": "", + "output": "A synonym for 'cold' is 'chilly'." + }, + { + "instruction": "Convert the mass from kilograms to grams.", + "input": "3 kilograms", + "output": "3 kilograms is equivalent to 3000 grams." + }, + { + "instruction": "What is the capital of Portugal?", + "input": "", + "output": "The capital of Portugal is Lisbon." + }, + { + "instruction": "Name a gas commonly found in the Earth's atmosphere.", + "input": "", + "output": "A gas commonly found in the Earth's atmosphere is nitrogen." + }, + { + "instruction": "Sort the following numbers in ascending order.", + "input": "5, 18, 3, 11.", + "output": "The numbers in ascending order would be: 3, 5, 11, 18." + }, + { + "instruction": "Generate a sentence using the word 'eloquent'.", + "input": "", + "output": "The speaker was eloquent and captivated the audience." + }, + { + "instruction": "Rewrite this sentence using appropriate grammar rules:", + "input": "he saw a old man", + "output": "He saw an old man." + }, + { + "instruction": "What is an antonym of 'conceal'?", + "input": "", + "output": "An antonym of 'conceal' is 'reveal'." + }, + { + "instruction": "Rewrite the following sentence so that it is in active voice.", + "input": "The cookies were eaten by the children.", + "output": "The children ate the cookies." + }, + { + "instruction": "What is the boiling point of water in Celsius?", + "input": "", + "output": "The boiling point of water is 100 degrees Celsius." + }, + { + "instruction": "Translate 'I love you' into Spanish.", + "input": "", + "output": "Te amo" + }, + { + "instruction": "Translate 'I am happy' into French.", + "input": "", + "output": "The French translation of 'I am happy' is 'Je suis heureux' (if the speaker is male) or 'Je suis heureuse' (if the speaker is female)." + }, + { + "instruction": "Generate an antonym for the word 'optimistic'.", + "input": "", + "output": "An antonym for the word 'optimistic' could be 'pessimistic'." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "Her smile is very bright.", + "output": "Her smile is a ray of sunshine." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'lovely'." + }, + { + "instruction": "Define the term 'photosynthesis'.", + "input": "", + "output": "Photosynthesis is the process by which green plants and some other organisms use sunlight to synthesize foods with the help of chlorophyll. It converts carbon dioxide and water into glucose and oxygen." + }, + { + "instruction": "Arrange the following events in chronological order: Invention of the airplane, Fall of the Berlin Wall, Discovery of America.", + "input": "", + "output": "1. Discovery of America\n2. Invention of the airplane\n3. Fall of the Berlin Wall" + }, + { + "instruction": "Determine the gravitational force acting on an object with a mass of 10 kg on Earth.", + "input": "", + "output": "The gravitational force acting on an object with a mass of 10 kg on Earth is approximately 98 newtons." + }, + { + "instruction": "Suggest a continuation for the sentence.", + "input": "When the sun set,", + "output": "\"the sky turned a brilliant shade of orange, casting a warm glow over the city.\"" + }, + { + "instruction": "Determine whether the given two words are synonyms or antonyms.", + "input": "Open - Close", + "output": "Open and Close are antonyms." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "She made a brief summary of the book.", + "output": "She summarized the book." + }, + { + "instruction": "Calculate the area of a triangle with a base of 10 units and height of 5 units.", + "input": "", + "output": "25 square units." + }, + { + "instruction": "Provide a synonym for 'funny'.", + "input": "", + "output": "A synonym for 'funny' is 'humorous'." + }, + { + "instruction": "Identify the theme of the story.", + "input": "The Tortoise and the Hare", + "output": "The theme of the story is that slow and steady wins the race." + }, + { + "instruction": "Arrange these historical periods in chronological order: Renaissance, Roman Empire, Victorian Era.", + "input": "", + "output": "1. Roman Empire\n2. Renaissance\n3. Victorian Era" + }, + { + "instruction": "Convert the following sentence to use a relative clause: The man is a teacher. He lives next door.", + "input": "", + "output": "The man who lives next door is a teacher." + }, + { + "instruction": "Name a dwarf planet in our solar system.", + "input": "", + "output": "One dwarf planet in our solar system is Pluto." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "He is very generous.", + "output": "He has a heart of gold." + }, + { + "instruction": "What is a synonym of 'brave'?", + "input": "", + "output": "A synonym for 'brave' is 'courageous'." + }, + { + "instruction": "Convert 100 pounds to kilograms.", + "input": "", + "output": "100 pounds is approximately 45.36 kilograms." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The sun was shining brightly.", + "output": "The sun was a golden coin in the sky." + }, + { + "instruction": "Identify the adjective in the sentence.", + "input": "The quick fox jumped over the lazy dog.", + "output": "The adjectives in the sentence are 'quick' and 'lazy'." + }, + { + "instruction": "Provide a synonym for 'strong'.", + "input": "", + "output": "robust." + }, + { + "instruction": "Generate a question with the keyword 'virus'", + "input": "", + "output": "How does a virus infect a cell?" + }, + { + "instruction": "Rewrite the following sentence using passive voice.", + "input": "The team achieved great results.", + "output": "Great results were achieved by the team." + }, + { + "instruction": "Translate 'hello' into Japanese.", + "input": "", + "output": "The Japanese translation of 'hello' is '\u3053\u3093\u306b\u3061\u306f' (Konnichiwa)." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "He is very wise.", + "output": "He is a fountain of wisdom." + }, + { + "instruction": "Generate a sentence using the word 'persistent'.", + "input": "", + "output": "He is persistent and never gives up on his goals." + }, + { + "instruction": "Generate a sentence that uses the word \u2018mystic\u2019.", + "input": "", + "output": "The ancient ruins had a mystic quality that captivated all who visited." + }, + { + "instruction": "What is a synonym for 'tiny'?", + "input": "", + "output": "A synonym for 'tiny' is 'small'." + }, + { + "instruction": "What is the capital of France?", + "input": "", + "output": "The capital of France is Paris." + }, + { + "instruction": "Generate a sentence using the word 'serene'.", + "input": "", + "output": "The lake was serene and calm in the early morning." + }, + { + "instruction": "Convert the phrase into a complete sentence.", + "input": "night sky", + "output": "The night sky was dotted with countless twinkling stars." + }, + { + "instruction": "Name the three branches of the United States government.", + "input": "", + "output": "The three branches of the United States government are the Executive, the Legislative, and the Judicial branches." + }, + { + "instruction": "Convert the angle from radians to degrees.", + "input": "\u03c0/2 radians", + "output": "\u03c0/2 radians is 90 degrees." + }, + { + "instruction": "Find the next 3 terms in the following sequence: 3, 5, 7, 9", + "input": "", + "output": "The next three terms in the sequence are: 11, 13, 15." + }, + { + "instruction": "Provide a synonym for 'pretty'.", + "input": "", + "output": "A synonym for 'pretty' is 'attractive'." + }, + { + "instruction": "Rewrite the following sentence to include a hyperbole: 'I'm very hungry.'", + "input": "I'm very hungry.", + "output": "I'm so hungry I could eat a horse." + }, + { + "instruction": "Scramble the given phrase.", + "input": "A sunny day", + "output": "A nnuys yda" + }, + { + "instruction": "Define the term 'economics'.", + "input": "", + "output": "Economics is a social science concerned with the production, distribution, and consumption of goods and services. It studies how individuals, businesses, governments, and nations make choices about how to allocate resources." + }, + { + "instruction": "What is an antonym of 'accelerate'?", + "input": "", + "output": "An antonym of 'accelerate' is 'decelerate'." + }, + { + "instruction": "Rewrite this sentence to start with \"Despite\": She managed to finish her work, although she was tired.", + "input": "", + "output": "Despite being tired, she managed to finish her work." + }, + { + "instruction": "Translate 'See you soon' into French.", + "input": "", + "output": "\u00c0 bient\u00f4t" + }, + { + "instruction": "What is the chemical formula for water?", + "input": "", + "output": "The chemical formula for water (dihydrogen monoxide) is H2O." + }, + { + "instruction": "Transform the following sentence into a command.", + "input": "You should clean your room.", + "output": "Clean your room." + }, + { + "instruction": "Find a synonym for the given verb.", + "input": "Speak", + "output": "Talk" + }, + { + "instruction": "What is the past tense of 'build'?", + "input": "", + "output": "The past tense of 'build' is 'built'." + }, + { + "instruction": "Translate 'Where is the nearest hospital?' into Italian.", + "input": "", + "output": "The Italian translation of 'Where is the nearest hospital?' is 'Dove \u00e8 l'ospedale pi\u00f9 vicino?'." + }, + { + "instruction": "Convert 200 meters to kilometers.", + "input": "", + "output": "200 meters is 0.2 kilometers." + }, + { + "instruction": "Determine the state of matter for water at 100 degrees Celsius.", + "input": "", + "output": "Water is in a gaseous state at 100 degrees Celsius." + }, + { + "instruction": "What is the opposite of 'weak'?", + "input": "", + "output": "The opposite of 'weak' is 'strong'." + }, + { + "instruction": "Define the term 'biodiversity'.", + "input": "", + "output": "Biodiversity refers to the variety of life in the world or in a particular habitat or ecosystem." + }, + { + "instruction": "Explain what a haiku is.", + "input": "", + "output": "A haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5." + }, + { + "instruction": "Name the process by which plants make their food.", + "input": "", + "output": "The process by which plants make their food is called photosynthesis." + }, + { + "instruction": "What is the melting point of iron?", + "input": "", + "output": "The melting point of iron is approximately 1538 degrees Celsius." + }, + { + "instruction": "Define the term 'evaporation'.", + "input": "", + "output": "Evaporation is the process by which water changes from a liquid to a gas or vapor." + }, + { + "instruction": "Define the concept of 'gravity'.", + "input": "", + "output": "Gravity is a force that attracts two bodies toward each other, the force that causes apples to fall toward the ground and planets to orbit stars." + }, + { + "instruction": "Define the literary term 'alliteration'.", + "input": "", + "output": "Alliteration is the occurrence of the same letter or sound at the beginning of adjacent or closely connected words." + }, + { + "instruction": "What is the pH of baking soda solution?", + "input": "", + "output": "The pH of a baking soda solution is typically around 9." + }, + { + "instruction": "What is an antonym of 'ascend'?", + "input": "", + "output": "An antonym of 'ascend' is 'descend'." + }, + { + "instruction": "Simplify the expression: 2(x + 3) + 4(x - 1).", + "input": "", + "output": "6x + 2." + }, + { + "instruction": "Calculate the circumference of a circle with a diameter of 10 cm.", + "input": "", + "output": "The circumference of the circle is approximately 31.42 cm." + }, + { + "instruction": "Generate a random sequence of 10 digits.", + "input": "", + "output": "Here is a randomly generated sequence of 10 digits: 9274036185." + }, + { + "instruction": "Insert a suitable adjective to complete the sentence.", + "input": "The movie was _____.", + "output": "The movie was thrilling." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The sky is very blue.", + "output": "The sky is a sapphire sea." + }, + { + "instruction": "What is the boiling point of water at sea level?", + "input": "", + "output": "The boiling point of water at sea level is 100 degrees Celsius." + }, + { + "instruction": "What is the cube of 5?", + "input": "", + "output": "The cube of 5 is 125." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very strong.", + "output": "He is as strong as an ox." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The house is very big.", + "output": "The house is as big as a castle." + }, + { + "instruction": "What is the capital of China?", + "input": "", + "output": "The capital of China is Beijing." + }, + { + "instruction": "List two synonyms for \"intelligent.\"", + "input": "", + "output": "Two synonyms for \"intelligent\" are \"smart\" and \"clever.\"" + }, + { + "instruction": "Explain the function of the liver in the human body.", + "input": "", + "output": "The liver functions to process nutrients absorbed from the digestive tract, detoxify chemicals, and metabolize drugs." + }, + { + "instruction": "Rewrite the sentence to improve clarity: 'He saw that the sky was clouds full.'", + "input": "He saw that the sky was clouds full.", + "output": "He saw that the sky was full of clouds." + }, + { + "instruction": "Rewrite the sentence using alliteration.", + "input": "The wind blew softly.", + "output": "The soft wind swept silently." + }, + { + "instruction": "What is an antonym for 'increase'?", + "input": "", + "output": "An antonym for 'increase' is 'decrease'." + }, + { + "instruction": "Classify this text as either a recipe or a novel.", + "input": "Combine flour, sugar, and eggs in a bowl.", + "output": "Recipe" + }, + { + "instruction": "rewrite this sentence as a question.", + "input": "Your brother went to London last year.", + "output": "Did your brother go to London last year?" + }, + { + "instruction": "Generate a sentence using the word 'ambitious'.", + "input": "", + "output": "She is ambitious and strives for success in all her endeavors." + }, + { + "instruction": "Classify each sentence as either declarative, interrogative, imperative, or exclamatory.", + "input": "Can you help me?", + "output": "The classification of the sentence \"Can you help me?\" is interrogative." + }, + { + "instruction": "Classify the following numbers as even or odd.", + "input": "2, 5, 18, 21", + "output": "Even numbers: 2, 18\nOdd numbers: 5, 21" + }, + { + "instruction": "What is the chemical formula for salt?", + "input": "", + "output": "The chemical formula for salt (sodium chloride) is NaCl." + }, + { + "instruction": "Find the temperature of freezing water under normal atmospheric pressure.", + "input": "", + "output": "Under normal atmospheric pressure, the temperature of freezing water is 0 degrees Celsius (32 degrees Fahrenheit)." + }, + { + "instruction": "Edit the following sentence to make it more formal.", + "input": "He's got to see it to believe it.", + "output": "He must see it to believe it." + }, + { + "instruction": "What is the opposite of 'easy'?", + "input": "", + "output": "The opposite of 'easy' is 'difficult'." + }, + { + "instruction": "Change the given verb to its past continuous form.", + "input": "eat", + "output": "Eating." + }, + { + "instruction": "Transform the following sentence into the passive voice.", + "input": "They will build a new park.", + "output": "A new park will be built by them." + }, + { + "instruction": "Evaluate this sentence for spelling and grammar mistakes.", + "input": "I beleive this is a grate idea.", + "output": "There are two errors in the sentence. The corrected sentence should be: 'I believe this is a great idea.'" + }, + { + "instruction": "What is the opposite of 'light'?", + "input": "", + "output": "The opposite of 'light' is 'dark'." + }, + { + "instruction": "Translate 'good morning' into French.", + "input": "", + "output": "The French translation of 'good morning' is 'bonjour'." + }, + { + "instruction": "What is an antonym of 'progressive'?", + "input": "", + "output": "An antonym of 'progressive' is 'regressive'." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'Her smile lights up the room.'", + "input": "", + "output": "Her smile is a beacon in the room." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Watermelon", + "output": "The fruit \"Watermelon\" is typically classified as a type of melon." + }, + { + "instruction": "What is the square root of 64?", + "input": "", + "output": "The square root of 64 is 8." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "He is very rich.", + "output": "He is rolling in dough." + }, + { + "instruction": "Classify the following animal as carnivore, herbivore, or omnivore.", + "input": "Deer", + "output": "Herbivore" + }, + { + "instruction": "Generate a sentence using the word 'serendipity'.", + "input": "", + "output": "Discovering the quaint caf\u00e9 down the alley was pure serendipity." + }, + { + "instruction": "Generate a sentence using the word 'resourceful'.", + "input": "", + "output": "He is very resourceful and can solve any problem." + }, + { + "instruction": "Classify the following item: hammer.", + "input": "", + "output": "A hammer can be classified as a tool." + }, + { + "instruction": "Convert the time '1500 hours' into standard 12-hour clock format.", + "input": "", + "output": "1500 hours is equivalent to 3:00 PM." + }, + { + "instruction": "Rewrite the sentence using a synonym.", + "input": "It's cold outside.", + "output": "It's chilly outside." + }, + { + "instruction": "Convert 6 liters to milliliters.", + "input": "", + "output": "6 liters is 6000 milliliters." + }, + { + "instruction": "What is the capital of Turkey?", + "input": "", + "output": "The capital of Turkey is Ankara." + }, + { + "instruction": "Change the following sentence to past continuous tense.", + "input": "They finish the game.", + "output": "They were finishing the game." + }, + { + "instruction": "Summarize the theme of 'The Great Gatsby'.", + "input": "", + "output": "The theme revolves around the American Dream and its disillusionment." + }, + { + "instruction": "Delete the redundant words in the following sentence.", + "input": "We went to see to the mountains.", + "output": "We went to see the mountains." + }, + { + "instruction": "Convert 5 kilograms to grams.", + "input": "", + "output": "5 kilograms is 5000 grams." + }, + { + "instruction": "Generate a random 8-digit number.", + "input": "", + "output": "Here is a randomly generated 8-digit number: 47382916." + }, + { + "instruction": "What is the opposite of 'quiet'?", + "input": "", + "output": "The opposite of 'quiet' is 'loud'." + }, + { + "instruction": "Translate the term 'sustainable development' into German.", + "input": "", + "output": "The German translation of 'sustainable development' is 'nachhaltige Entwicklung'." + }, + { + "instruction": "Convert the following statement into an exclamation.", + "input": "This is amazing", + "output": "This is amazing!" + }, + { + "instruction": "Classify the item as either a grain or a fruit.", + "input": "Wheat", + "output": "Grain" + }, + { + "instruction": "What is the boiling point of water under standard conditions?", + "input": "", + "output": "100 degrees Celsius." + }, + { + "instruction": "Identify the tense used in the sentence.", + "input": "They had watched the movie.", + "output": "The tense used in the sentence 'They had watched the movie' is the past perfect tense." + }, + { + "instruction": "Rewrite the following sentence to include a personification.", + "input": "The wind is very strong today.", + "output": "The wind howls with might today, as if declaring its presence." + }, + { + "instruction": "Cut the following sentence into two, keeping the original meaning intact.", + "input": "She baked a cake and then she decorated it.", + "output": "She baked a cake. Then she decorated it." + }, + { + "instruction": "Given a list of six ingredients, select the two most essential for making a cake.", + "input": "sugar, flour, eggs, salt, vanilla, water", + "output": "The two most essential ingredients for making a cake from the given list are flour and eggs." + }, + { + "instruction": "What is the contraction for 'they will'?", + "input": "", + "output": "The contraction for 'they will' is 'they'll.'" + }, + { + "instruction": "Translate 'peace' into French.", + "input": "", + "output": "The French word for 'peace' is 'paix'." + }, + { + "instruction": "Classify the item as either a fruit or a vegetable.", + "input": "Tomato", + "output": "Fruit" + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'Her laughter is like music.'", + "input": "", + "output": "Her laughter is music to the ears." + }, + { + "instruction": "Capitalize each word in the sentence.", + "input": "the book is on the table", + "output": "The Book Is On The Table" + }, + { + "instruction": "Rewrite the following sentence to use a simile: The car raced down the road.", + "input": "", + "output": "The car raced down the road like a bullet." + }, + { + "instruction": "Combine the two ideas in the sentence.", + "input": "Running is good for health. It can be tiring.", + "output": "Running is good for health but can be tiring." + }, + { + "instruction": "Identify the main subject in the sentence.", + "input": "The excited students ran to the playground.", + "output": "The main subject in the sentence is 'The excited students.'" + }, + { + "instruction": "Name a synonym for 'angry.'", + "input": "", + "output": "A synonym for 'angry' is 'furious.'" + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She studied hard. She passed the exam.", + "output": "She studied hard, so she passed the exam." + }, + { + "instruction": "Explain the process of combustion.", + "input": "", + "output": "Combustion is a chemical process in which a substance reacts rapidly with oxygen, releasing heat and light. It typically produces carbon dioxide, water, and energy." + }, + { + "instruction": "Convert the given sentence into active voice.", + "input": "The proposal was approved by the committee.", + "output": "The committee approved the proposal." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The car is very fast.", + "output": "The car is as fast as lightning." + }, + { + "instruction": "What type of cloud is typically associated with thunderstorms?", + "input": "", + "output": "The type of cloud typically associated with thunderstorms is cumulonimbus." + }, + { + "instruction": "Name the author of 'Pride and Prejudice'.", + "input": "", + "output": "Jane Austen." + }, + { + "instruction": "What is the periodic symbol for chlorine?", + "input": "", + "output": "The periodic symbol for chlorine is Cl." + }, + { + "instruction": "Correct the punctuation in the sentence.", + "input": "Its time to go home.", + "output": "The corrected sentence should be: 'It's time to go home.'" + }, + { + "instruction": "Rewrite the sentence.", + "input": "The lecture was delivered in a clear manner.", + "output": "The lecture was delivered clearly." + }, + { + "instruction": "Generate a humorous anecdote.", + "input": "", + "output": "Why was the math book sad? Because it had too many problems!" + }, + { + "instruction": "Identify the correct spelling: 'recieve' or 'receive'.", + "input": "", + "output": "The correct spelling is 'receive'." + }, + { + "instruction": "Create a sentence using the word 'nostalgia'.", + "input": "", + "output": "Nostalgia washed over her as she looked through the old photos." + }, + { + "instruction": "Classify the following numbers as prime or composite.", + "input": ": 11, 14, 19.", + "output": "Prime numbers: 11, 19\nComposite numbers: 14" + }, + { + "instruction": "What is the capital of Denmark?", + "input": "", + "output": "The capital of Denmark is Copenhagen." + }, + { + "instruction": "What is the opposite of 'wet'?", + "input": "", + "output": "The opposite of 'wet' is 'dry'." + }, + { + "instruction": "Identify the type of sentence.", + "input": "Did you finish the report?", + "output": "The type of sentence is interrogative." + }, + { + "instruction": "What is the chemical symbol for mercury?", + "input": "", + "output": "Hg." + }, + { + "instruction": "Describe the process of active transport in cells.", + "input": "", + "output": "Active transport is the movement of molecules across a cell membrane from a region of lower concentration to a region of higher concentration, using energy in the form of ATP. This process is crucial for maintaining cellular homeostasis." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Mercury, oxygen, wood", + "output": "Mercury - Liquid\nOxygen - Gas\nWood - Solid" + }, + { + "instruction": "Convert 3 kilometers to meters.", + "input": "", + "output": "3 kilometers is 3000 meters." + }, + { + "instruction": "Convert the sentence to use an indefinite pronoun.", + "input": "Someone left a note.", + "output": "A note was left by someone." + }, + { + "instruction": "Provide a synonym for 'excited'.", + "input": "", + "output": "A synonym for 'excited' is 'thrilled'." + }, + { + "instruction": "Generate a sentence that follows the pattern: \"Never have I ever _____ without _____\"", + "input": "", + "output": "Never have I ever traveled without a map." + }, + { + "instruction": "Pick out the adjective from the following list.", + "input": "run, tall, quickly", + "output": "The correct adjective from the list is 'tall.'" + }, + { + "instruction": "Convert 1000 grams to kilograms.", + "input": "", + "output": "1000 grams is equal to 1 kilogram." + }, + { + "instruction": "What is the opposite of 'deep'?", + "input": "", + "output": "The opposite of 'deep' is 'shallow'." + }, + { + "instruction": "Categorize the given list of animals.", + "input": "Shark, Dolphin, Trout", + "output": "Fish: Shark, Trout\nMammals: Dolphin" + }, + { + "instruction": "Translate 'library' into Spanish.", + "input": "", + "output": "The Spanish word for 'library' is 'biblioteca'." + }, + { + "instruction": "Define the term 'hyperbole'.", + "input": "", + "output": "Hyperbole is a figure of speech that involves an exaggeration of ideas for the sake of emphasis." + }, + { + "instruction": "Translate 'Hello' into Russian.", + "input": "", + "output": "\u041f\u0440\u0438\u0432\u0435\u0442 (Privet)" + }, + { + "instruction": "Define the term 'kinetic energy'.", + "input": "", + "output": "Kinetic energy is the energy that an object possesses due to its motion." + }, + { + "instruction": "What is the opposite of 'hot'?", + "input": "", + "output": "The opposite of 'hot' is 'cold'." + }, + { + "instruction": "Convert 5 miles to kilometers.", + "input": "", + "output": "5 miles is approximately 8.05 kilometers." + }, + { + "instruction": "What is the chemical formula for magnesium sulfate?", + "input": "", + "output": "The chemical formula for magnesium sulfate is MgSO4." + }, + { + "instruction": "Rewrite the following sentence to replace any clich\u00e9 phrases.", + "input": "It's a piece of cake", + "output": "It's very easy." + }, + { + "instruction": "List five different types of vegetables.", + "input": "", + "output": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato\n5. Spinach" + }, + { + "instruction": "Convert 7 kilometers to meters.", + "input": "", + "output": "7 kilometers is 7000 meters." + }, + { + "instruction": "What is the opposite of 'heavy'?", + "input": "", + "output": "The opposite of 'heavy' is 'light'." + }, + { + "instruction": "What is the past tense of 'sing'?", + "input": "", + "output": "The past tense of 'sing' is 'sang.'" + }, + { + "instruction": "What is the molecular formula for carbon dioxide?", + "input": "", + "output": "The molecular formula for carbon dioxide is CO2." + }, + { + "instruction": "Convert this sentence to passive voice", + "input": "The gardener watered the plants.", + "output": "The plants were watered by the gardener." + }, + { + "instruction": "What is the past tense of 'throw'?", + "input": "", + "output": "The past tense of 'throw' is 'threw'." + }, + { + "instruction": "Explain what a sonnet is.", + "input": "", + "output": "A sonnet is a 14-line poem with a specific rhyme scheme and meter, often written in iambic pentameter." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "The company is known for its innovative products." + }, + { + "instruction": "Provide the plural form of 'cactus'.", + "input": "", + "output": "The plural form of 'cactus' is 'cacti'." + }, + { + "instruction": "Translate the phrase 'Where is the bathroom?' into German.", + "input": "", + "output": "The German translation of 'Where is the bathroom?' is 'Wo ist die Toilette?'" + }, + { + "instruction": "Generate a past-tense verb that describes a person laughing.", + "input": "", + "output": "Laughed." + }, + { + "instruction": "Generate a sentence using the word 'transient.'", + "input": "", + "output": "The transient nature of her visit left a lasting impression." + }, + { + "instruction": "Generate a sentence using the word 'optimistic'.", + "input": "", + "output": "He remained optimistic despite the challenges he faced." + }, + { + "instruction": "Re-word this sentence using an indirect question.", + "input": "What time is the meeting?", + "output": "Could you tell me what time the meeting is?" + }, + { + "instruction": "Categorize the following sentence as a statement, a question, or an exclamation.", + "input": "What a beautiful day!", + "output": "Exclamation." + }, + { + "instruction": "What is the opposite of 'rich'?", + "input": "", + "output": "The opposite of 'rich' is 'poor'." + }, + { + "instruction": "Find a synonym for the given verb.", + "input": "Begin", + "output": "Commence" + }, + { + "instruction": "Edit the given text to ensure all plural nouns are spelled correctly.", + "input": "The birds sings beautiful songs.", + "output": "The birds sing beautiful songs." + }, + { + "instruction": "Transform the following sentence into a question using \"could.\"", + "input": "You can help me tomorrow.", + "output": "Could you help me tomorrow?" + }, + { + "instruction": "Classify the following items: bicycle, rose, tiger.", + "input": "", + "output": "Vehicles: Bicycle\nPlants: Rose\nAnimals: Tiger" + }, + { + "instruction": "Define the term 'irony'.", + "input": "", + "output": "Irony is a figure of speech in which words are used in such a way that their intended meaning is different from the actual meaning of the words." + }, + { + "instruction": "Translate 'Welcome' into German.", + "input": "", + "output": "The German translation of 'Welcome' is 'Willkommen'." + }, + { + "instruction": "Explain the primary function of the human heart.", + "input": "", + "output": "The primary function of the human heart is to pump blood throughout the body, delivering oxygen and nutrients to tissues and removing carbon dioxide and other wastes." + }, + { + "instruction": "Reword the following sentence to the future tense.", + "input": "He is reading a novel inspired by his grandmother.", + "output": "He will be reading a novel inspired by his grandmother." + }, + { + "instruction": "Convert the given sentence into active voice.", + "input": "The law was passed by the government.", + "output": "The government passed the law." + }, + { + "instruction": "Create a sentence using the word 'inevitable'.", + "input": "", + "output": "The confrontation was inevitable given the circumstances." + }, + { + "instruction": "Categorize the following sentence as either factual or opinion-based.", + "input": "Chocolate is the best dessert.", + "output": "Opinion-based." + }, + { + "instruction": "What is an antonym of 'old'?", + "input": "", + "output": "young." + }, + { + "instruction": "Provide a synonym for 'hardworking'.", + "input": "", + "output": "A synonym for 'hardworking' is 'diligent'." + }, + { + "instruction": "What is the boiling point of sulfur in Celsius?", + "input": "", + "output": "The boiling point of sulfur is 444.6 degrees Celsius." + }, + { + "instruction": "What is the plural form of 'child'?", + "input": "", + "output": "The plural form of 'child' is 'children'." + }, + { + "instruction": "What is an antonym of 'complicated'?", + "input": "", + "output": "An antonym of 'complicated' is 'simple'." + }, + { + "instruction": "Name three forms of water.", + "input": "", + "output": "The three forms of water are solid (ice), liquid (water), and gas (steam)." + }, + { + "instruction": "Rewrite this sentence as a question.", + "input": "The dog chased the cat.", + "output": "Did the dog chase the cat?" + }, + { + "instruction": "Split the following sentence into two declarative sentences: 'The movie was long but interesting.'", + "input": "", + "output": "The movie was long. It was interesting." + }, + { + "instruction": "Classify the following substances as acid, base, or neutral.", + "input": "Lemon juice, Soap, Water", + "output": "Acid: Lemon juice\nBase: Soap\nNeutral: Water" + }, + { + "instruction": "What is a synonym of 'sad'?", + "input": "", + "output": "A synonym for 'sad' is 'unhappy'." + }, + { + "instruction": "Correct any spelling mistakes in the given sentence.", + "input": "I prefer homemade cookies to store boaght.", + "output": "I prefer homemade cookies to store bought." + }, + { + "instruction": "Generate a sentence using the word 'transient'.", + "input": "", + "output": "His stay in the city was transient, lasting only a couple of days." + }, + { + "instruction": "Translate 'I am lost' into Italian.", + "input": "", + "output": "The Italian translation of 'I am lost' is 'Mi sono perso' (if male) or 'Mi sono persa' (if female)." + }, + { + "instruction": "Classify this text as a technical document or a narrative.", + "input": "This manual provides instructions for installing the software.", + "output": "Technical document" + }, + { + "instruction": "Sort the following list in descending order.", + "input": "10, 2, 25, 16, 7", + "output": "25, 16, 10, 7, 2." + }, + { + "instruction": "Translate 'Can I have some water?' into French.", + "input": "", + "output": "Puis-je avoir de l'eau?" + }, + { + "instruction": "Create a simile with the word 'as cold as'.", + "input": "", + "output": "Her hands were as cold as ice." + }, + { + "instruction": "Classify the following words by their grammatical categories: swim, beautiful, quickly", + "input": "", + "output": "Swim: Verb\nBeautiful: Adjective\nQuickly: Adverb" + }, + { + "instruction": "Calculate the density of an object with a mass of 15 grams and a volume of 5 cubic centimeters.", + "input": "", + "output": "The density of the object is 3 grams per cubic centimeter." + }, + { + "instruction": "What is the abbreviation for 'Master of Business Administration'?", + "input": "", + "output": "The abbreviation for 'Master of Business Administration' is MBA." + }, + { + "instruction": "Convert the following number from Roman numerals: IX.", + "input": "", + "output": "The number IX in Roman numerals is 9." + }, + { + "instruction": "What is the opposite of 'horizontal'?", + "input": "", + "output": "The opposite of 'horizontal' is 'vertical'." + }, + { + "instruction": "Translate 'Where can I buy tickets?' into Italian.", + "input": "", + "output": "The Italian translation for 'Where can I buy tickets?' is 'Dove posso comprare i biglietti?'" + }, + { + "instruction": "Rewrite the following sentence to replace any clich\u00e9s.", + "input": "He was as cool as a cucumber.", + "output": "He remained very calm." + }, + { + "instruction": "Identify the main verb in the sentence.", + "input": "The dog barked loudly.", + "output": "The main verb in the sentence is 'barked'." + }, + { + "instruction": "Generate a sentence using the word 'elucidate'.", + "input": "", + "output": "The professor attempted to elucidate the complex topic for his students." + }, + { + "instruction": "Correct the sentence.", + "input": "Me and my friend went to the store.", + "output": "My friend and I went to the store." + }, + { + "instruction": "What is the formula for calculating work done?", + "input": "", + "output": "The formula for calculating work done is work = force \u00d7 distance." + }, + { + "instruction": "What is the chemical formula for ammonium nitrate?", + "input": "", + "output": "The chemical formula for ammonium nitrate is NH4NO3." + }, + { + "instruction": "What is the molecular formula for water?", + "input": "", + "output": "The molecular formula for water is H2O." + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The food was not good.", + "output": "The food could use some improvement." + }, + { + "instruction": "What is the opposite of 'lazy'?", + "input": "", + "output": "The opposite of 'lazy' is 'diligent'." + }, + { + "instruction": "Name three essential vitamins for human health.", + "input": "", + "output": "1. Vitamin A\n2. Vitamin C\n3. Vitamin D" + }, + { + "instruction": "Define the term 'simile'.", + "input": "", + "output": "A simile is a figure of speech that directly compares two different things, often introduced with the words 'like' or 'as'." + }, + { + "instruction": "What is the boiling point of chlorine in Celsius?", + "input": "", + "output": "The boiling point of chlorine is -34 degrees Celsius." + }, + { + "instruction": "Translate 'My name is' into French.", + "input": "", + "output": "The French translation of 'My name is' is 'Je m'appelle'." + }, + { + "instruction": "Convert 200 centimeters to meters.", + "input": "", + "output": "200 centimeters is 2 meters." + }, + { + "instruction": "What is the chemical symbol for zinc?", + "input": "", + "output": "Zn." + }, + { + "instruction": "What is the formula for calculating force?", + "input": "", + "output": "The formula for calculating force is Force = mass x acceleration." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "The company's innovative approach set it apart from its competitors." + }, + { + "instruction": "Generate a sentence using the word 'generous'.", + "input": "", + "output": "He is very generous and always helps those in need." + }, + { + "instruction": "What is an antonym of 'sharp'?", + "input": "", + "output": "An antonym of 'sharp' is 'dull'." + }, + { + "instruction": "Describe the basic structure of a neuron.", + "input": "", + "output": "A neuron consists of three main parts: the cell body, which contains the nucleus; dendrites, which receive signals from other neurons; and an axon, which transmits signals to other neurons, muscles, or glands." + }, + { + "instruction": "Translate 'see you later' into Spanish.", + "input": "", + "output": "Hasta luego" + }, + { + "instruction": "What are the first 10 elements of the Fibonacci sequence?", + "input": "", + "output": "0, 1, 1, 2, 3, 5, 8, 13, 21, 34." + }, + { + "instruction": "What is an antonym of 'transparent'?", + "input": "", + "output": "An antonym of 'transparent' is 'opaque'." + }, + { + "instruction": "What is the past tense of 'think'?", + "input": "", + "output": "The past tense of 'think' is 'thought'." + }, + { + "instruction": "Classify each sentence as either declarative, interrogative, imperative, or exclamatory.", + "input": "Please open the door.", + "output": "The classification of the sentence 'Please open the door.' is imperative." + }, + { + "instruction": "Rewrite the sentence to use a negative adverb.", + "input": "She always remembers to call.", + "output": "She never forgets to call." + }, + { + "instruction": "Convert 50 miles per hour to kilometers per hour.", + "input": "", + "output": "50 miles per hour is approximately 80.47 kilometers per hour." + }, + { + "instruction": "Convert the active sentence to passive: 'The chef cooks the meal every day.'", + "input": "", + "output": "The meal is cooked by the chef every day." + }, + { + "instruction": "Classify an input string as either a noun or a verb.", + "input": "Dance", + "output": "'Dance' can be classified as a verb." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The book is very interesting.", + "output": "The book is a page-turner." + }, + { + "instruction": "Generate a question with the keyword \u201cbacteria\u201d", + "input": "", + "output": "How do bacteria affect human health?" + }, + { + "instruction": "What is the past participle of 'run'?", + "input": "", + "output": "The past participle of 'run' is 'run'." + }, + { + "instruction": "Provide a synonym for 'quick'.", + "input": "", + "output": "A synonym for 'quick' is 'rapid'." + }, + { + "instruction": "Give an example of a metaphor.", + "input": "", + "output": "An example of a metaphor is saying, 'Time is a thief,' which suggests that time, like a thief, can take away life moments without literally stealing." + }, + { + "instruction": "Calculate the interest on a $1,000 investment at an annual rate of 5% over 3 years.", + "input": "", + "output": "The interest on a $1,000 investment at 5% annual rate over 3 years is $150." + }, + { + "instruction": "What is a synonym of 'elated'?", + "input": "", + "output": "A synonym for 'elated' is 'overjoyed'." + }, + { + "instruction": "Translate 'It's raining' into Japanese.", + "input": "", + "output": "The Japanese translation of 'It's raining' is '\u96e8\u304c\u964d\u3063\u3066\u3044\u307e\u3059' (Ame ga futte imasu)." + }, + { + "instruction": "Convert the following verb to its past participle form: 'speak'", + "input": "speak", + "output": "The past participle form of 'speak' is 'spoken.'" + }, + { + "instruction": "Describe the principle of conservation of energy.", + "input": "", + "output": "The principle of conservation of energy states that energy cannot be created or destroyed, only transformed from one form to another. This means the total energy of an isolated system remains constant over time." + }, + { + "instruction": "Calculate the perimeter of a rectangle with a length of 5 meters and a width of 3 meters.", + "input": "", + "output": "The perimeter of the rectangle is 16 meters." + }, + { + "instruction": "What is the past tense of 'freeze'?", + "input": "", + "output": "The past tense of 'freeze' is 'froze'." + }, + { + "instruction": "Correct the misuse of 'their' in the sentence.", + "input": "Their going to the store.", + "output": "The corrected sentence should be: 'They're going to the store.'" + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer crashed.", + "output": "The statement \"My computer crashed\" can be classified as negative." + }, + { + "instruction": "Identify the onomatopoeia in the sentence.", + "input": "The bees buzzed as they flew past.", + "output": "The onomatopoeia in the sentence is 'buzzed'." + }, + { + "instruction": "Edit the sentence.", + "input": "We enjoys watching movies.", + "output": "We enjoy watching movies." + }, + { + "instruction": "Define the term 'gravity'.", + "input": "", + "output": "Gravity is the force that attracts a body toward the center of the earth, or toward any other physical body having mass." + }, + { + "instruction": "What is an antonym for 'hot'?", + "input": "", + "output": "An antonym for 'hot' is 'cold'." + }, + { + "instruction": "Reverse the order of the given phrase.", + "input": "sun and moon", + "output": "moon and sun" + }, + { + "instruction": "What is the formula for calculating power?", + "input": "", + "output": "The formula for calculating power is power = work/time." + }, + { + "instruction": "Name the process by which plants absorb water through their roots.", + "input": "", + "output": "The process by which plants absorb water through their roots is called absorption." + }, + { + "instruction": "What is the past tense of 'draw'?", + "input": "", + "output": "drew." + }, + { + "instruction": "Change the given verb to its present perfect form.", + "input": "run", + "output": "Run." + }, + { + "instruction": "Define the term 'oxymoron'.", + "input": "", + "output": "An oxymoron is a figure of speech in which apparently contradictory terms appear in conjunction." + }, + { + "instruction": "What is the chemical symbol for water?", + "input": "", + "output": "The chemical symbol for water is H2O." + }, + { + "instruction": "Convert the active sentence to passive: 'They paint the house every year.'", + "input": "", + "output": "The house is painted every year by them." + }, + { + "instruction": "Provide the past tense form of the verb \"run.\"", + "input": "", + "output": "The past tense of the verb \"run\" is \"ran.\"" + }, + { + "instruction": "Convert the sentence to passive voice: 'They will hold the meeting tomorrow.'", + "input": "They will hold the meeting tomorrow.", + "output": "The meeting will be held tomorrow by them." + }, + { + "instruction": "What is the chemical formula for sulfuric acid?", + "input": "", + "output": "The chemical formula for sulfuric acid is H2SO4." + }, + { + "instruction": "What type of cloud indicates fair weather?", + "input": "", + "output": "Cumulus clouds often indicate fair weather." + }, + { + "instruction": "Insert a suitable adjective to complete the sentence.", + "input": "The journey was _____.", + "output": "The journey was arduous." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "The test was very easy.", + "output": "The test was a piece of cake." + }, + { + "instruction": "What is the opposite of 'increase'?", + "input": "", + "output": "The opposite of 'increase' is 'decrease'." + }, + { + "instruction": "Calculate the perimeter of a square with a side length of 6 meters.", + "input": "", + "output": "The perimeter of the square is 24 meters." + }, + { + "instruction": "Find the missing number in the sequence", + "input": "2, 4, ?, 8, 10", + "output": "The missing number in the sequence is 6." + }, + { + "instruction": "Translate 'Good night' into German.", + "input": "", + "output": "The German translation of 'Good night' is 'Gute Nacht'." + }, + { + "instruction": "Provide a word that rhymes with 'light.'", + "input": "", + "output": "A word that rhymes with 'light' is 'might.'" + }, + { + "instruction": "What is the comparative form of 'good'?", + "input": "", + "output": "The comparative form of 'good' is 'better'." + }, + { + "instruction": "Name the literary device used in 'The wind whispered through the trees.'", + "input": "", + "output": "The literary device used is personification." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'His words cut deeper than a knife.'", + "input": "", + "output": "His words were a knife, cutting deep." + }, + { + "instruction": "Generate a sentence using the word 'innovate'.", + "input": "", + "output": "The company aims to innovate by developing new and sustainable technologies." + }, + { + "instruction": "What is the speed of light?", + "input": "", + "output": "The speed of light is approximately 299,792 kilometers per second." + }, + { + "instruction": "Identify the adverb in the sentence", + "input": "She quickly ran to the store.", + "output": "The adverb in the sentence is 'quickly'." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'gorgeous'." + }, + { + "instruction": "What is the capital of India?", + "input": "", + "output": "The capital of India is New Delhi." + }, + { + "instruction": "Classify the following words by their grammatical categories: jump, quick, beautiful", + "input": "", + "output": "Jump: Verb\nQuick: Adjective\nBeautiful: Adjective" + }, + { + "instruction": "What is the chemical formula for potassium nitrate?", + "input": "", + "output": "The chemical formula for potassium nitrate is KNO3." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The cake is very sweet.", + "output": "The cake is as sweet as honey." + }, + { + "instruction": "Label whether the following statements are true or false.", + "input": "The moon is a planet.", + "output": "False" + }, + { + "instruction": "Convert 1000 milliliters to liters.", + "input": "", + "output": "1000 milliliters is equivalent to 1 liter." + }, + { + "instruction": "Change this sentence into the past continuous tense.", + "input": "She sings in the choir.", + "output": "She was singing in the choir." + }, + { + "instruction": "Provide a synonym for 'clever.'", + "input": "", + "output": "A synonym for 'clever' is 'smart.'" + }, + { + "instruction": "Change the sentence 'You should have called me.' into a question.", + "input": "", + "output": "Should you have called me?" + } +] \ No newline at end of file diff --git a/Codes/ch07/01_main-chapter-code/load-finetuned-model.ipynb b/Codes/ch07/01_main-chapter-code/load-finetuned-model.ipynb new file mode 100644 index 0000000..8420841 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/load-finetuned-model.ipynb @@ -0,0 +1,219 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "1545a16b-bc8d-4e49-b9a6-db6631e7483d", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "f3f83194-82b9-4478-9550-5ad793467bd0", + "metadata": {}, + "source": [ + "# Load And Use Finetuned Model" + ] + }, + { + "cell_type": "markdown", + "id": "466b564e-4fd5-4d76-a3a1-63f9f0993b7e", + "metadata": {}, + "source": [ + "This notebook contains minimal code to load the finetuned model that was instruction finetuned and saved in chapter 7 via [ch07.ipynb](ch07.ipynb)." + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "fd80e5f5-0f79-4a6c-bf31-2026e7d30e52", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tiktoken version: 0.7.0\n", + "torch version: 2.4.0\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\n", + " \"tiktoken\", # Tokenizer\n", + " \"torch\", # Deep learning library\n", + "]\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "ed86d6b7-f32d-4601-b585-a2ea3dbf7201", + "metadata": {}, + "outputs": [], + "source": [ + "from pathlib import Path\n", + "\n", + "finetuned_model_path = Path(\"gpt2-medium355M-sft.pth\")\n", + "if not finetuned_model_path.exists():\n", + " print(\n", + " f\"Could not find '{finetuned_model_path}'.\\n\"\n", + " \"Run the `ch07.ipynb` notebook to finetune and save the finetuned model.\"\n", + " )" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "fb02584a-5e31-45d5-8377-794876907bc6", + "metadata": {}, + "outputs": [], + "source": [ + "from previous_chapters import GPTModel\n", + "\n", + "\n", + "BASE_CONFIG = {\n", + " \"vocab_size\": 50257, # Vocabulary size\n", + " \"context_length\": 1024, # Context length\n", + " \"drop_rate\": 0.0, # Dropout rate\n", + " \"qkv_bias\": True # Query-key-value bias\n", + "}\n", + "\n", + "model_configs = {\n", + " \"gpt2-small (124M)\": {\"emb_dim\": 768, \"n_layers\": 12, \"n_heads\": 12},\n", + " \"gpt2-medium (355M)\": {\"emb_dim\": 1024, \"n_layers\": 24, \"n_heads\": 16},\n", + " \"gpt2-large (774M)\": {\"emb_dim\": 1280, \"n_layers\": 36, \"n_heads\": 20},\n", + " \"gpt2-xl (1558M)\": {\"emb_dim\": 1600, \"n_layers\": 48, \"n_heads\": 25},\n", + "}\n", + "\n", + "CHOOSE_MODEL = \"gpt2-medium (355M)\"\n", + "\n", + "BASE_CONFIG.update(model_configs[CHOOSE_MODEL])\n", + "\n", + "model_size = CHOOSE_MODEL.split(\" \")[-1].lstrip(\"(\").rstrip(\")\")\n", + "model = GPTModel(BASE_CONFIG)" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "f1ccf2b7-176e-4cfd-af7a-53fb76010b94", + "metadata": {}, + "outputs": [], + "source": [ + "import torch\n", + "\n", + "model.load_state_dict(torch.load(\n", + " \"gpt2-medium355M-sft.pth\",\n", + " map_location=torch.device(\"cpu\"),\n", + " weights_only=True\n", + "))\n", + "model.eval();" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "a1fd174e-9555-46c5-8780-19b0aa4f26e5", + "metadata": {}, + "outputs": [], + "source": [ + "import tiktoken\n", + "\n", + "tokenizer = tiktoken.get_encoding(\"gpt2\")" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "2a4c0129-efe5-46e9-bb90-ba08d407c1a2", + "metadata": {}, + "outputs": [], + "source": [ + "prompt = \"\"\"Below is an instruction that describes a task. Write a response \n", + "that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Convert the active sentence to passive: 'The chef cooks the meal every day.'\n", + "\"\"\"" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "1e26862c-10b5-4a0f-9dd6-b6ddbad2fc3f", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "The meal is cooked every day by the chef.\n" + ] + } + ], + "source": [ + "from previous_chapters import (\n", + " generate,\n", + " text_to_token_ids,\n", + " token_ids_to_text\n", + ")\n", + "\n", + "def extract_response(response_text, input_text):\n", + " return response_text[len(input_text):].replace(\"### Response:\", \"\").strip()\n", + "\n", + "torch.manual_seed(123)\n", + "\n", + "token_ids = generate(\n", + " model=model,\n", + " idx=text_to_token_ids(prompt, tokenizer),\n", + " max_new_tokens=35,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + ")\n", + "\n", + "response = token_ids_to_text(token_ids, tokenizer)\n", + "response = extract_response(response, prompt)\n", + "print(response)" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.4" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/01_main-chapter-code/ollama_evaluate.py b/Codes/ch07/01_main-chapter-code/ollama_evaluate.py new file mode 100644 index 0000000..e2e0204 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/ollama_evaluate.py @@ -0,0 +1,126 @@ +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch +# +# A minimal instruction finetuning file based on the code in chapter 7 + +import json +import psutil +from tqdm import tqdm +import urllib.request + + +def query_model(prompt, model="llama3", url="http://localhost:11434/api/chat"): + # Create the data payload as a dictionary + data = { + "model": model, + "messages": [ + {"role": "user", "content": prompt} + ], + "options": { # Settings below are required for deterministic responses + "seed": 123, + "temperature": 0, + "num_ctx": 2048 + } + } + + # Convert the dictionary to a JSON formatted string and encode it to bytes + payload = json.dumps(data).encode("utf-8") + + # Create a request object, setting the method to POST and adding necessary headers + request = urllib.request.Request(url, data=payload, method="POST") + request.add_header("Content-Type", "application/json") + + # Send the request and capture the response + response_data = "" + with urllib.request.urlopen(request) as response: + # Read and decode the response + while True: + line = response.readline().decode("utf-8") + if not line: + break + response_json = json.loads(line) + response_data += response_json["message"]["content"] + + return response_data + + +def check_if_running(process_name): + running = False + for proc in psutil.process_iter(["name"]): + if process_name in proc.info["name"]: + running = True + break + return running + + +def format_input(entry): + instruction_text = ( + f"Below is an instruction that describes a task. " + f"Write a response that appropriately completes the request." + f"\n\n### Instruction:\n{entry['instruction']}" + ) + + input_text = f"\n\n### Input:\n{entry['input']}" if entry["input"] else "" + + return instruction_text + input_text + + +def main(file_path): + ollama_running = check_if_running("ollama") + + if not ollama_running: + raise RuntimeError("Ollama not running. Launch ollama before proceeding.") + print("Ollama running:", check_if_running("ollama")) + + with open(file_path, "r") as file: + test_data = json.load(file) + + model = "llama3" + scores = generate_model_scores(test_data, "model_response", model) + print(f"Number of scores: {len(scores)} of {len(test_data)}") + print(f"Average score: {sum(scores)/len(scores):.2f}\n") + + +def generate_model_scores(json_data, json_key, model="llama3"): + scores = [] + for entry in tqdm(json_data, desc="Scoring entries"): + if entry[json_key] == "": + scores.append(0) + else: + prompt = ( + f"Given the input `{format_input(entry)}` " + f"and correct output `{entry['output']}`, " + f"score the model response `{entry[json_key]}`" + f" on a scale from 0 to 100, where 100 is the best score. " + f"Respond with the integer number only." + ) + score = query_model(prompt, model) + try: + scores.append(int(score)) + except ValueError: + print(f"Could not convert score: {score}") + continue + + return scores + + +if __name__ == "__main__": + + import argparse + + parser = argparse.ArgumentParser( + description="Evaluate model responses with ollama" + ) + parser.add_argument( + "--file_path", + required=True, + help=( + "The path to the test dataset `.json` file with the" + " `'output'` and `'model_response'` keys" + ) + ) + args = parser.parse_args() + + main(file_path=args.file_path) diff --git a/Codes/ch07/01_main-chapter-code/previous_chapters.py b/Codes/ch07/01_main-chapter-code/previous_chapters.py new file mode 100644 index 0000000..090eab5 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/previous_chapters.py @@ -0,0 +1,470 @@ +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch +# +# This file collects all the relevant code that we covered thus far +# throughout Chapters 2-6. +# This file can be run as a standalone script. + + +import matplotlib.pyplot as plt +from matplotlib.ticker import MaxNLocator +import numpy as np +import tiktoken +import torch +import torch.nn as nn +from torch.utils.data import Dataset, DataLoader + + +##################################### +# Chapter 2 +##################################### + + +class GPTDatasetV1(Dataset): + def __init__(self, txt, tokenizer, max_length, stride): + self.tokenizer = tokenizer + self.input_ids = [] + self.target_ids = [] + + # Tokenize the entire text + token_ids = tokenizer.encode(txt, allowed_special={"<|endoftext|>"}) + + # Use a sliding window to chunk the book into overlapping sequences of max_length + for i in range(0, len(token_ids) - max_length, stride): + input_chunk = token_ids[i:i + max_length] + target_chunk = token_ids[i + 1: i + max_length + 1] + self.input_ids.append(torch.tensor(input_chunk)) + self.target_ids.append(torch.tensor(target_chunk)) + + def __len__(self): + return len(self.input_ids) + + def __getitem__(self, idx): + return self.input_ids[idx], self.target_ids[idx] + + +def create_dataloader_v1(txt, batch_size=4, max_length=256, + stride=128, shuffle=True, drop_last=True, num_workers=0): + # Initialize the tokenizer + tokenizer = tiktoken.get_encoding("gpt2") + + # Create dataset + dataset = GPTDatasetV1(txt, tokenizer, max_length, stride) + + # Create dataloader + dataloader = DataLoader( + dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=num_workers) + + return dataloader + + +##################################### +# Chapter 3 +##################################### +class MultiHeadAttention(nn.Module): + def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False): + super().__init__() + assert d_out % num_heads == 0, "d_out must be divisible by n_heads" + + self.d_out = d_out + self.num_heads = num_heads + self.head_dim = d_out // num_heads # Reduce the projection dim to match desired output dim + + self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias) + self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias) + self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias) + self.out_proj = nn.Linear(d_out, d_out) # Linear layer to combine head outputs + self.dropout = nn.Dropout(dropout) + self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1)) + + def forward(self, x): + b, num_tokens, d_in = x.shape + + keys = self.W_key(x) # Shape: (b, num_tokens, d_out) + queries = self.W_query(x) + values = self.W_value(x) + + # We implicitly split the matrix by adding a `num_heads` dimension + # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim) + keys = keys.view(b, num_tokens, self.num_heads, self.head_dim) + values = values.view(b, num_tokens, self.num_heads, self.head_dim) + queries = queries.view(b, num_tokens, self.num_heads, self.head_dim) + + # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim) + keys = keys.transpose(1, 2) + queries = queries.transpose(1, 2) + values = values.transpose(1, 2) + + # Compute scaled dot-product attention (aka self-attention) with a causal mask + attn_scores = queries @ keys.transpose(2, 3) # Dot product for each head + + # Original mask truncated to the number of tokens and converted to boolean + mask_bool = self.mask.bool()[:num_tokens, :num_tokens] + + # Use the mask to fill attention scores + attn_scores.masked_fill_(mask_bool, -torch.inf) + + attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1) + attn_weights = self.dropout(attn_weights) + + # Shape: (b, num_tokens, num_heads, head_dim) + context_vec = (attn_weights @ values).transpose(1, 2) + + # Combine heads, where self.d_out = self.num_heads * self.head_dim + context_vec = context_vec.reshape(b, num_tokens, self.d_out) + context_vec = self.out_proj(context_vec) # optional projection + + return context_vec + + +##################################### +# Chapter 4 +##################################### +class LayerNorm(nn.Module): + def __init__(self, emb_dim): + super().__init__() + self.eps = 1e-5 + self.scale = nn.Parameter(torch.ones(emb_dim)) + self.shift = nn.Parameter(torch.zeros(emb_dim)) + + def forward(self, x): + mean = x.mean(dim=-1, keepdim=True) + var = x.var(dim=-1, keepdim=True, unbiased=False) + norm_x = (x - mean) / torch.sqrt(var + self.eps) + return self.scale * norm_x + self.shift + + +class GELU(nn.Module): + def __init__(self): + super().__init__() + + def forward(self, x): + return 0.5 * x * (1 + torch.tanh( + torch.sqrt(torch.tensor(2.0 / torch.pi)) * + (x + 0.044715 * torch.pow(x, 3)) + )) + + +class FeedForward(nn.Module): + def __init__(self, cfg): + super().__init__() + self.layers = nn.Sequential( + nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]), + GELU(), + nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]), + ) + + def forward(self, x): + return self.layers(x) + + +class TransformerBlock(nn.Module): + def __init__(self, cfg): + super().__init__() + self.att = MultiHeadAttention( + d_in=cfg["emb_dim"], + d_out=cfg["emb_dim"], + context_length=cfg["context_length"], + num_heads=cfg["n_heads"], + dropout=cfg["drop_rate"], + qkv_bias=cfg["qkv_bias"]) + self.ff = FeedForward(cfg) + self.norm1 = LayerNorm(cfg["emb_dim"]) + self.norm2 = LayerNorm(cfg["emb_dim"]) + self.drop_resid = nn.Dropout(cfg["drop_rate"]) + + def forward(self, x): + # Shortcut connection for attention block + shortcut = x + x = self.norm1(x) + x = self.att(x) # Shape [batch_size, num_tokens, emb_size] + x = self.drop_resid(x) + x = x + shortcut # Add the original input back + + # Shortcut connection for feed-forward block + shortcut = x + x = self.norm2(x) + x = self.ff(x) + x = self.drop_resid(x) + x = x + shortcut # Add the original input back + + return x + + +class GPTModel(nn.Module): + def __init__(self, cfg): + super().__init__() + self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"]) + self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"]) + self.drop_emb = nn.Dropout(cfg["drop_rate"]) + + self.trf_blocks = nn.Sequential( + *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])]) + + self.final_norm = LayerNorm(cfg["emb_dim"]) + self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False) + + def forward(self, in_idx): + batch_size, seq_len = in_idx.shape + tok_embeds = self.tok_emb(in_idx) + pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device)) + x = tok_embeds + pos_embeds # Shape [batch_size, num_tokens, emb_size] + x = self.drop_emb(x) + x = self.trf_blocks(x) + x = self.final_norm(x) + logits = self.out_head(x) + return logits + + +def generate_text_simple(model, idx, max_new_tokens, context_size): + # idx is (B, T) array of indices in the current context + for _ in range(max_new_tokens): + + # Crop current context if it exceeds the supported context size + # E.g., if LLM supports only 5 tokens, and the context size is 10 + # then only the last 5 tokens are used as context + idx_cond = idx[:, -context_size:] + + # Get the predictions + with torch.no_grad(): + logits = model(idx_cond) + + # Focus only on the last time step + # (batch, n_token, vocab_size) becomes (batch, vocab_size) + logits = logits[:, -1, :] + + # Get the idx of the vocab entry with the highest logits value + idx_next = torch.argmax(logits, dim=-1, keepdim=True) # (batch, 1) + + # Append sampled index to the running sequence + idx = torch.cat((idx, idx_next), dim=1) # (batch, n_tokens+1) + + return idx + + +##################################### +# Chapter 5 +##################################### +def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None): + + # For-loop is the same as before: Get logits, and only focus on last time step + for _ in range(max_new_tokens): + idx_cond = idx[:, -context_size:] + with torch.no_grad(): + logits = model(idx_cond) + logits = logits[:, -1, :] + + # New: Filter logits with top_k sampling + if top_k is not None: + # Keep only top_k values + top_logits, _ = torch.topk(logits, top_k) + min_val = top_logits[:, -1] + logits = torch.where(logits < min_val, torch.tensor(float('-inf')).to(logits.device), logits) + + # New: Apply temperature scaling + if temperature > 0.0: + logits = logits / temperature + + # Apply softmax to get probabilities + probs = torch.softmax(logits, dim=-1) # (batch_size, context_len) + + # Sample from the distribution + idx_next = torch.multinomial(probs, num_samples=1) # (batch_size, 1) + + # Otherwise same as before: get idx of the vocab entry with the highest logits value + else: + idx_next = torch.argmax(logits, dim=-1, keepdim=True) # (batch_size, 1) + + if idx_next == eos_id: # Stop generating early if end-of-sequence token is encountered and eos_id is specified + break + + # Same as before: append sampled index to the running sequence + idx = torch.cat((idx, idx_next), dim=1) # (batch_size, num_tokens+1) + + return idx + + +def train_model_simple(model, train_loader, val_loader, optimizer, device, num_epochs, + eval_freq, eval_iter, start_context, tokenizer): + # Initialize lists to track losses and tokens seen + train_losses, val_losses, track_tokens_seen = [], [], [] + tokens_seen, global_step = 0, -1 + + # Main training loop + for epoch in range(num_epochs): + model.train() # Set model to training mode + + for input_batch, target_batch in train_loader: + optimizer.zero_grad() # Reset loss gradients from previous batch iteration + loss = calc_loss_batch(input_batch, target_batch, model, device) + loss.backward() # Calculate loss gradients + optimizer.step() # Update model weights using loss gradients + tokens_seen += input_batch.numel() + global_step += 1 + + # Optional evaluation step + if global_step % eval_freq == 0: + train_loss, val_loss = evaluate_model( + model, train_loader, val_loader, device, eval_iter) + train_losses.append(train_loss) + val_losses.append(val_loss) + track_tokens_seen.append(tokens_seen) + print(f"Ep {epoch+1} (Step {global_step:06d}): " + f"Train loss {train_loss:.3f}, Val loss {val_loss:.3f}") + + # Print a sample text after each epoch + generate_and_print_sample( + model, tokenizer, device, start_context + ) + + return train_losses, val_losses, track_tokens_seen + + +def evaluate_model(model, train_loader, val_loader, device, eval_iter): + model.eval() + with torch.no_grad(): + train_loss = calc_loss_loader(train_loader, model, device, num_batches=eval_iter) + val_loss = calc_loss_loader(val_loader, model, device, num_batches=eval_iter) + model.train() + return train_loss, val_loss + + +def generate_and_print_sample(model, tokenizer, device, start_context): + model.eval() + context_size = model.pos_emb.weight.shape[0] + encoded = text_to_token_ids(start_context, tokenizer).to(device) + with torch.no_grad(): + token_ids = generate_text_simple( + model=model, idx=encoded, + max_new_tokens=50, context_size=context_size + ) + decoded_text = token_ids_to_text(token_ids, tokenizer) + print(decoded_text.replace("\n", " ")) # Compact print format + model.train() + + +def assign(left, right): + if left.shape != right.shape: + raise ValueError(f"Shape mismatch. Left: {left.shape}, Right: {right.shape}") + return torch.nn.Parameter(torch.tensor(right)) + + +def load_weights_into_gpt(gpt, params): + gpt.pos_emb.weight = assign(gpt.pos_emb.weight, params['wpe']) + gpt.tok_emb.weight = assign(gpt.tok_emb.weight, params['wte']) + + for b in range(len(params["blocks"])): + q_w, k_w, v_w = np.split( + (params["blocks"][b]["attn"]["c_attn"])["w"], 3, axis=-1) + gpt.trf_blocks[b].att.W_query.weight = assign( + gpt.trf_blocks[b].att.W_query.weight, q_w.T) + gpt.trf_blocks[b].att.W_key.weight = assign( + gpt.trf_blocks[b].att.W_key.weight, k_w.T) + gpt.trf_blocks[b].att.W_value.weight = assign( + gpt.trf_blocks[b].att.W_value.weight, v_w.T) + + q_b, k_b, v_b = np.split( + (params["blocks"][b]["attn"]["c_attn"])["b"], 3, axis=-1) + gpt.trf_blocks[b].att.W_query.bias = assign( + gpt.trf_blocks[b].att.W_query.bias, q_b) + gpt.trf_blocks[b].att.W_key.bias = assign( + gpt.trf_blocks[b].att.W_key.bias, k_b) + gpt.trf_blocks[b].att.W_value.bias = assign( + gpt.trf_blocks[b].att.W_value.bias, v_b) + + gpt.trf_blocks[b].att.out_proj.weight = assign( + gpt.trf_blocks[b].att.out_proj.weight, + params["blocks"][b]["attn"]["c_proj"]["w"].T) + gpt.trf_blocks[b].att.out_proj.bias = assign( + gpt.trf_blocks[b].att.out_proj.bias, + params["blocks"][b]["attn"]["c_proj"]["b"]) + + gpt.trf_blocks[b].ff.layers[0].weight = assign( + gpt.trf_blocks[b].ff.layers[0].weight, + params["blocks"][b]["mlp"]["c_fc"]["w"].T) + gpt.trf_blocks[b].ff.layers[0].bias = assign( + gpt.trf_blocks[b].ff.layers[0].bias, + params["blocks"][b]["mlp"]["c_fc"]["b"]) + gpt.trf_blocks[b].ff.layers[2].weight = assign( + gpt.trf_blocks[b].ff.layers[2].weight, + params["blocks"][b]["mlp"]["c_proj"]["w"].T) + gpt.trf_blocks[b].ff.layers[2].bias = assign( + gpt.trf_blocks[b].ff.layers[2].bias, + params["blocks"][b]["mlp"]["c_proj"]["b"]) + + gpt.trf_blocks[b].norm1.scale = assign( + gpt.trf_blocks[b].norm1.scale, + params["blocks"][b]["ln_1"]["g"]) + gpt.trf_blocks[b].norm1.shift = assign( + gpt.trf_blocks[b].norm1.shift, + params["blocks"][b]["ln_1"]["b"]) + gpt.trf_blocks[b].norm2.scale = assign( + gpt.trf_blocks[b].norm2.scale, + params["blocks"][b]["ln_2"]["g"]) + gpt.trf_blocks[b].norm2.shift = assign( + gpt.trf_blocks[b].norm2.shift, + params["blocks"][b]["ln_2"]["b"]) + + gpt.final_norm.scale = assign(gpt.final_norm.scale, params["g"]) + gpt.final_norm.shift = assign(gpt.final_norm.shift, params["b"]) + gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"]) + + +def text_to_token_ids(text, tokenizer): + encoded = tokenizer.encode(text, allowed_special={"<|endoftext|>"}) + encoded_tensor = torch.tensor(encoded).unsqueeze(0) # add batch dimension + return encoded_tensor + + +def token_ids_to_text(token_ids, tokenizer): + flat = token_ids.squeeze(0) # remove batch dimension + return tokenizer.decode(flat.tolist()) + + +def calc_loss_batch(input_batch, target_batch, model, device): + input_batch, target_batch = input_batch.to(device), target_batch.to(device) + logits = model(input_batch) + loss = torch.nn.functional.cross_entropy(logits.flatten(0, 1), target_batch.flatten()) + return loss + + +def calc_loss_loader(data_loader, model, device, num_batches=None): + total_loss = 0. + if len(data_loader) == 0: + return float("nan") + elif num_batches is None: + num_batches = len(data_loader) + else: + # Reduce the number of batches to match the total number of batches in the data loader + # if num_batches exceeds the number of batches in the data loader + num_batches = min(num_batches, len(data_loader)) + for i, (input_batch, target_batch) in enumerate(data_loader): + if i < num_batches: + loss = calc_loss_batch(input_batch, target_batch, model, device) + total_loss += loss.item() + else: + break + return total_loss / num_batches + + +def plot_losses(epochs_seen, tokens_seen, train_losses, val_losses): + fig, ax1 = plt.subplots(figsize=(5, 3)) + + # Plot training and validation loss against epochs + ax1.plot(epochs_seen, train_losses, label="Training loss") + ax1.plot(epochs_seen, val_losses, linestyle="-.", label="Validation loss") + ax1.set_xlabel("Epochs") + ax1.set_ylabel("Loss") + ax1.legend(loc="upper right") + ax1.xaxis.set_major_locator(MaxNLocator(integer=True)) # only show integer labels on x-axis + + # Create a second x-axis for tokens seen + ax2 = ax1.twiny() # Create a second x-axis that shares the same y-axis + ax2.plot(tokens_seen, train_losses, alpha=0) # Invisible plot for aligning ticks + ax2.set_xlabel("Tokens seen") + + fig.tight_layout() # Adjust layout to make room + plt.savefig("loss-plot.pdf") + plt.show() diff --git a/Codes/ch07/01_main-chapter-code/tests.py b/Codes/ch07/01_main-chapter-code/tests.py new file mode 100644 index 0000000..40ee892 --- /dev/null +++ b/Codes/ch07/01_main-chapter-code/tests.py @@ -0,0 +1,16 @@ +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch + +# File for internal use (unit tests) + + +import subprocess + + +def test_gpt_class_finetune(): + command = ["python", "ch06/01_main-chapter-code/gpt_class_finetune.py", "--test_mode"] + + result = subprocess.run(command, capture_output=True, text=True) + assert result.returncode == 0, f"Script exited with errors: {result.stderr}" diff --git a/Codes/ch07/02_dataset-utilities/README.md b/Codes/ch07/02_dataset-utilities/README.md new file mode 100644 index 0000000..2a6b8aa --- /dev/null +++ b/Codes/ch07/02_dataset-utilities/README.md @@ -0,0 +1,80 @@ +# Chapter 7: Finetuning to Follow Instructions + +This folder contains utility code that can be used for preparing an instruction dataset. + +Install the additional package requirements via: + +```bash +pip install -r requirements-extra.txt +``` + + + + + +### Finding Near Duplicates + +The `find-near-duplicates.py` function can be used to identify duplicates and near-duplicates in an instruction dataset. For example, + + + +```bash +python find-near-duplicates.py --json_file instruction-examples.json +``` + +``` +scikit-learn version: 1.3.1 + + +================================================== +Searching 'instruction' for duplicates ... +================================================== +Duplicate pair found with similarity 0.94: +1. Edit the following sentence to make it more formal. +2. Edit the sentence to make it more formal. + +Duplicate pair found with similarity 1.00: +1. Name a dwarf planet in our solar system. +2. Name a dwarf planet in our solar system. + +Duplicate pair found with similarity 0.91: +1. Change the sentences from active voice to passive voice. +2. Change the sentence from passive to active voice. + + + +================================================== +Searching 'input' for duplicates ... +================================================== +No duplicates found + + +================================================== +Searching 'output' for duplicates ... +================================================== +Duplicate pair found with similarity 1.00: +1. One dwarf planet in our solar system is Pluto. +2. One dwarf planet in our solar system is Pluto. + + +``` + +  +You can use the `--threshold` setting with a value between 0 and 1 to decrease or increase the sensitivity. +The default threshold is 0.9. + + + +  + ## Creating Passive Voice Entries + + - The [create-passive-voice-entries.ipynb](create-passive-voice-entries.ipynb) notebook uses OpenAI's GPT-4 to create "passive voice" entries for an instruction dataset, as shown in the example below + + ```python + { + 'instruction': 'Identify the verb in the following sentence', + 'input': 'The cat sleeps on the couch.', + 'output': 'The verb in the sentence is "sleeps."', + 'output_2': 'The sentence is "sleeps."' # <---- Newly created entry + } + ``` diff --git a/Codes/ch07/02_dataset-utilities/config.json b/Codes/ch07/02_dataset-utilities/config.json new file mode 100644 index 0000000..a01cb47 --- /dev/null +++ b/Codes/ch07/02_dataset-utilities/config.json @@ -0,0 +1,4 @@ +{ + "OPENAI_API_KEY": "sk-...", + "_comment": "Enter your API key from https://platform.openai.com/api-keys" +} diff --git a/Codes/ch07/02_dataset-utilities/create-passive-voice-entries.ipynb b/Codes/ch07/02_dataset-utilities/create-passive-voice-entries.ipynb new file mode 100644 index 0000000..c5029e6 --- /dev/null +++ b/Codes/ch07/02_dataset-utilities/create-passive-voice-entries.ipynb @@ -0,0 +1,426 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "136a4efe-fb99-4311-8679-e0a5b6282755", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "b1910a06-e8a3-40ac-8201-ff70615b1ba4", + "metadata": { + "tags": [] + }, + "source": [ + "# Create \"Passive Voice\" Entries for an Instruction Dataset" + ] + }, + { + "cell_type": "markdown", + "id": "a128651b-f326-4232-a994-42f38b7ed520", + "metadata": {}, + "source": [ + "- This notebook uses OpenAI's GPT-4 to create \"passive voice\" entries for an instruction dataset, as shown in the example below\n", + "\n", + "```python\n", + "{ \n", + " 'instruction': 'Identify the verb in the following sentence',\n", + " 'input': 'The cat sleeps on the couch.',\n", + " 'output': 'The verb in the sentence is \"sleeps.\"',\n", + " 'output_2': 'The sentence is \"sleeps.\"' # <---- Newly created entry\n", + "} \n", + "```" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "267ba0d1-b884-42df-85bd-0be746fd47a5", + "metadata": {}, + "outputs": [], + "source": [ + "# pip install -r requirements-extra.txt" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "63610acc-db94-437f-8d38-e99dca0299cb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "openai version: 1.30.3\n", + "tqdm version: 4.65.0\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\"openai\", # OpenAI API\n", + " \"tqdm\", # Progress bar\n", + " ]\n", + "\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "markdown", + "id": "8bcdcb34-ac75-4f4f-9505-3ce0666c42d5", + "metadata": {}, + "source": [ + "## Test OpenAI API" + ] + }, + { + "cell_type": "markdown", + "id": "9558a522-650d-401a-84fc-9fd7b1f39da7", + "metadata": {}, + "source": [ + "- First, let's test if the OpenAI API is correctly set up\n", + "- If you don't have an account yet, you need to create one at https://platform.openai.com/\n", + "- Note that you will also have to transfer some funds to your account as the GPT-4 API is not free (see https://platform.openai.com/settings/organization/billing/overview)\n", + "- Creating the ~200 passive voice entries using the code in this notebook costs about $0.13 (13 cents)" + ] + }, + { + "cell_type": "markdown", + "id": "89343a84-0ddc-42fc-bf50-298a342b93c0", + "metadata": {}, + "source": [ + "- First, we need to provide our OpenAI API secret key, which can be found at https://platform.openai.com/api-keys\n", + "- Make sure not to share this key with anyone\n", + "- Add this secret key (`\"sk-...\"`) to the `config.json` file in this folder" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "26900564-aba7-48ba-8ee8-6cc9a505a25c", + "metadata": {}, + "outputs": [], + "source": [ + "import json\n", + "from openai import OpenAI\n", + "\n", + "# Load API key from a JSON file. \n", + "# Make sure to replace \"sk-...\" with your actual API key from https://platform.openai.com/api-keys\n", + "with open(\"config.json\", \"r\") as config_file:\n", + " config = json.load(config_file)\n", + " api_key = config[\"OPENAI_API_KEY\"]\n", + "\n", + "client = OpenAI(api_key=api_key)" + ] + }, + { + "cell_type": "markdown", + "id": "16642a48-1cab-40d2-af08-ab8c2fbf5876", + "metadata": {}, + "source": [ + "- First, let's try the API with a simple example to make sure it works as intended:" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "08e9ef2e-e816-4283-840e-43625791ad33", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'Breakfast was eaten by me.'" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "def run_chatgpt(prompt, client, model=\"gpt-4-turbo\"):\n", + " response = client.chat.completions.create(\n", + " model=model,\n", + " messages=[{\"role\": \"user\", \"content\": prompt}],\n", + " temperature=0.0,\n", + " )\n", + " return response.choices[0].message.content\n", + "\n", + "\n", + "# Prepare input\n", + "sentence = \"I ate breakfast\"\n", + "prompt = f\"Convert the following sentence to passive voice: '{sentence}'\"\n", + "run_chatgpt(prompt, client)" + ] + }, + { + "cell_type": "markdown", + "id": "162a4739-6f03-4092-a5c2-f57a0b6a4c4d", + "metadata": {}, + "source": [ + "## Create JSON Entries" + ] + }, + { + "cell_type": "markdown", + "id": "ca011a8b-20c5-4101-979e-9b5fccf62f8a", + "metadata": {}, + "source": [ + "- Next, we load the file we want to modify:" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "8b2d393a-aa92-4190-9d44-44326a6f699b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of entries: 200\n" + ] + } + ], + "source": [ + "import json\n", + "\n", + "json_file = \"instruction-examples.json\"\n", + "\n", + "with open(json_file, \"r\") as file:\n", + " json_data = json.load(file)\n", + " \n", + "print(\"Number of entries:\", len(json_data))" + ] + }, + { + "cell_type": "markdown", + "id": "39a55283-7d51-4136-ba60-f799d49f4098", + "metadata": {}, + "source": [ + "- And we try the OpenAI chat API on a small sample first to ensure that it works correctly:" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "735cc089-d127-480a-b39d-0782581f0c41", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Input:\n", + ">> The verb in the sentence is \"sleeps.\"\n", + "\n", + "Output:\n", + ">> The sentence is \"sleeps.\"\n", + "\n", + "-------------------------\n", + "\n", + "Input:\n", + ">> The plural form of \"goose\" is \"geese.\"\n", + "\n", + "Output:\n", + ">> The plural form of \"goose\" is referred to as \"geese.\"\n", + "\n", + "-------------------------\n", + "\n", + "Input:\n", + ">> The three primary colors are red, blue, and yellow.\n", + "\n", + "Output:\n", + ">> Red, blue, and yellow are considered the three primary colors.\n", + "\n", + "-------------------------\n", + "\n", + "Input:\n", + ">> They had finished the game.\n", + "\n", + "Output:\n", + ">> The game had been finished by them.\n", + "\n", + "-------------------------\n", + "\n", + "Input:\n", + ">> The abbreviation for \"Doctor of Philosophy\" is Ph.D.\n", + "\n", + "Output:\n", + ">> The abbreviation \"Ph.D.\" is used for \"Doctor of Philosophy\".\n", + "\n", + "-------------------------\n" + ] + } + ], + "source": [ + "for entry in json_data[:5]:\n", + " text = entry[\"output\"]\n", + " prompt = f\"Without adding any response or explanation, convert the following text to passive voice: {text}\"\n", + " \n", + " print(\"\\nInput:\")\n", + " print(\">>\", text)\n", + " print(\"\\nOutput:\")\n", + " print(\">>\", run_chatgpt(prompt, client))\n", + " print(\"\\n-------------------------\")" + ] + }, + { + "cell_type": "markdown", + "id": "142dfaa7-429f-4eb0-b74d-ff327f79547a", + "metadata": {}, + "source": [ + "- Let's now extend the code to add the generated entries to the `json_data` and add a progress bar:" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "4f700d4b-19e5-4404-afa7-b0f093024232", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|██████████████████████████████████████████████████████████████████████| 5/5 [00:04<00:00, 1.23it/s]\n" + ] + } + ], + "source": [ + "from tqdm import tqdm # a progress bar tool\n", + "\n", + "\n", + "for i, entry in tqdm(enumerate(json_data[:5]), total=len(json_data[:5])):\n", + " text = entry[\"output\"]\n", + " prompt = f\"Without adding any response or explanation, convert the following text to passive voice: {text}\"\n", + " json_data[i][\"output_2\"] = run_chatgpt(prompt, client)" + ] + }, + { + "cell_type": "markdown", + "id": "cd144282-0596-4e9b-9815-322cff34b400", + "metadata": {}, + "source": [ + "- One more time, let's make sure that the new entries (`\"output_2\"`) look ok" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "5b6eaa87-a86d-42a1-a20a-b764b0d559d4", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'instruction': 'Identify the verb in the following sentence: The cat sleeps on the couch.',\n", + " 'input': '',\n", + " 'output': 'The verb in the sentence is \"sleeps.\"',\n", + " 'output_2': 'The sentence is \"sleeps.\"'}" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "json_data[0]" + ] + }, + { + "cell_type": "markdown", + "id": "6970e8cf-2b18-4e3d-9f25-e6a4489c39a7", + "metadata": {}, + "source": [ + "- Finally, if everything above looks ok, let's run the conversion to passive voice on our entire json dataset (this takes about 3 minutes):" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "eef99407-8ffd-4a63-b7ab-ffe30c0f0677", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|██████████████████████████████████████████████████████████████████| 200/200 [03:43<00:00, 1.12s/it]\n" + ] + } + ], + "source": [ + "for i, entry in tqdm(enumerate(json_data), total=len(json_data)):\n", + " text = entry[\"output\"]\n", + " prompt = f\"Without adding any response or explanation, convert the following text to passive voice: {text}\"\n", + " json_data[i][\"output_2\"] = run_chatgpt(prompt, client)" + ] + }, + { + "cell_type": "markdown", + "id": "ac91ae85-2f0e-456a-be1d-56e1958f30d8", + "metadata": {}, + "source": [ + "- After the conversion is completed, we save the file:" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "330cc30a-b08e-4bf0-bee2-bec0da4208de", + "metadata": {}, + "outputs": [], + "source": [ + "new_json_file = json_file.replace(\".json\", \"-modified.json\")\n", + "\n", + "\n", + "with open(new_json_file, \"w\") as file:\n", + " json.dump(json_data, file, indent=4) # \"indent\" for pretty-printing" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.6" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/02_dataset-utilities/find-near-duplicates.py b/Codes/ch07/02_dataset-utilities/find-near-duplicates.py new file mode 100644 index 0000000..45b2fce --- /dev/null +++ b/Codes/ch07/02_dataset-utilities/find-near-duplicates.py @@ -0,0 +1,153 @@ + +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch + +import argparse +import json +import re +from sklearn import __version__ as sklearn_version +from sklearn.feature_extraction.text import TfidfVectorizer +from sklearn.metrics.pairwise import cosine_similarity + + +# Sample JSON dataset +example_data = [ + {"instruction": "What is the capital of Italy?", + "input": "", "output": "The capital of Italy is Rome." + }, + {"instruction": "What's the capital city of Italy?", + "input": "", "output": "The capital city is Rome." + }, + {"instruction": "Identify the main verb in the sentence: 'The cat sleeps on the couch.'", + "input": "", "output": "The verb is 'sleeps'." + }, + {"instruction": "Identify the verb in the following sentence: The cat sleeps on the couch.", + "input": "", "output": "The verb in the sentence is \"sleeps.\"" + }, + # ... +] + + +def preprocess_text(text): + # Lowercase the text + text = text.lower() + # Remove punctuation + text = re.sub(r'[^\w\s]', '', text) + return text + + +def find_near_duplicates(json_data, threshold=0.75, key="instruction"): + """The higher the threshold, the more similar the texts have to be to match""" + + # Extract instructions + text = [preprocess_text(item[key]) for item in json_data if item[key]] + near_duplicates = [] + indices_to_remove = set() + + if not text: + return {}, near_duplicates + + # Vectorize the text data + vectorizer = TfidfVectorizer(stop_words=None, analyzer='char', ngram_range=(1, 3)) + tfidf_matrix = vectorizer.fit_transform(text) + + # Compute cosine similarity between each pair of entries + cos_sim_matrix = cosine_similarity(tfidf_matrix) + + # Find pairs of near-duplicate instructions based on the threshold + + for i in range(len(cos_sim_matrix)): + for j in range(i+1, len(cos_sim_matrix)): + if cos_sim_matrix[i, j] > threshold: + if len(json_data[i][key]) <= 1 or len(json_data[j][key]) <= 1: + continue + near_duplicates.append((json_data[i], json_data[j], cos_sim_matrix[i, j])) + if key in ("input", "output"): # Don't remove duplicates based on the instruction + indices_to_remove.add(j) # Mark the second entry for removal + + # Remove the near-duplicate entries + filtered_json_data = [item for index, item in enumerate(json_data) if index not in indices_to_remove] + + return filtered_json_data, near_duplicates + + +def find_print_and_remove_near_duplicates(json_data, remove_duplicates=False, threshold=0.75): + """ + Searches each key in the first JSON object for duplicates across a list of JSON objects. + Prints the duplicates if found. + """ + for key in json_data[0].keys(): + + if remove_duplicates: + json_data, near_duplicates = find_near_duplicates(json_data, key=key, threshold=threshold) + else: + _, near_duplicates = find_near_duplicates(json_data, key=key, threshold=threshold) + separator = 50 * '=' + print(f"\n\n{separator}\nSearching '{key}' for duplicates ...\n{separator}") + if not near_duplicates: + print("No duplicates found") + else: + for dup in near_duplicates: + print( + f"Duplicate pair found with similarity {dup[2]:.2f}:\n" + f"1. {dup[0][key]}\n2. {dup[1][key]}\n" + ) + return json_data + + +if __name__ == "__main__": + print("scikit-learn version:", sklearn_version) + + parser = argparse.ArgumentParser() + parser.add_argument( + "--json_file", + type=str, + help=("Path to the dataset JSON file") + ) + parser.add_argument( + "--threshold", + type=float, + default=0.9, + help=("A sensitivity threshold between 0 and 1 where 1 is strictest") + ) + parser.add_argument( + "--remove_duplicates", + action='store_true', + default=False, + help=( + "Removes duplicates based on the 'input' or 'output' keys " + " (but not the 'instruction') and saves the cleaned JSON file as --json_output_file" + ) + ) + parser.add_argument( + "--json_output_file", + type=str, + help=("Path to the dataset JSON file") + ) + + args = parser.parse_args() + + if args.remove_duplicates and not args.json_output_file: + raise ValueError( + "Provide an output file via --json_output_file " + "to save the cleaned JSON data." + ) + + if not args.json_file: + json_data = example_data + + else: + with open(args.json_file, "r") as file: + json_data = json.load(file) + + json_data = find_print_and_remove_near_duplicates( + json_data=json_data, + remove_duplicates=args.remove_duplicates, + threshold=args.threshold + ) + + if args.remove_duplicates: + with open(args.json_output_file, "w") as file: + json.dump(json_data, file, indent=4) diff --git a/Codes/ch07/02_dataset-utilities/instruction-examples-modified.json b/Codes/ch07/02_dataset-utilities/instruction-examples-modified.json new file mode 100644 index 0000000..333b83b --- /dev/null +++ b/Codes/ch07/02_dataset-utilities/instruction-examples-modified.json @@ -0,0 +1,1202 @@ +[ + { + "instruction": "Identify the verb in the following sentence: The cat sleeps on the couch.", + "input": "", + "output": "The verb in the sentence is \"sleeps.\"", + "output_2": "The sentence is \"The verb in the sentence is slept.\"" + }, + { + "instruction": "What is the plural form of \"goose\"?", + "input": "", + "output": "The plural form of \"goose\" is \"geese.\"", + "output_2": "The plural form of \"goose\" is referred to as \"geese.\"" + }, + { + "instruction": "List three primary colors.", + "input": "", + "output": "The three primary colors are red, blue, and yellow.", + "output_2": "Red, blue, and yellow are the three primary colors." + }, + { + "instruction": "Change the following sentence to past perfect tense.", + "input": "They finish the game.", + "output": "They had finished the game.", + "output_2": "The game had been finished by them." + }, + { + "instruction": "What is the abbreviation for \"Doctor of Philosophy\"?", + "input": "", + "output": "The abbreviation for \"Doctor of Philosophy\" is Ph.D.", + "output_2": "The abbreviation \"Ph.D.\" is used for \"Doctor of Philosophy\"." + }, + { + "instruction": "Correct the misspelled word in the following sentence: I recieved your message.", + "input": "", + "output": "The misspelled word \"recieved\" should be spelled \"received.\"", + "output_2": "The misspelled word \"recieved\" should be spelled \"received\" should be stated." + }, + { + "instruction": "Transform the following sentence to include an exclamation mark to show excitement.", + "input": "I won the lottery", + "output": "I won the lottery!", + "output_2": "The lottery was won by me!" + }, + { + "instruction": "What is the antonym of \"ancient\"?", + "input": "", + "output": "The antonym of \"ancient\" is \"modern.\"", + "output_2": "The antonym of \"ancient\" is said to be \"modern.\"" + }, + { + "instruction": "Name a synonym for \"happiness.\"", + "input": "", + "output": "A synonym for \"happiness\" is \"joy.\"", + "output_2": "A synonym for \"happiness\" is referred to as \"joy.\"" + }, + { + "instruction": "Classify the following items: gold ring, daisy, python.", + "input": "", + "output": "Animals: Python\nPlants: Daisy\nMinerals: Gold ring", + "output_2": "Animals: Python is possessed.\nPlants: Daisy is grown.\nMinerals: Gold ring is made." + }, + { + "instruction": "Generate three adjectives to describe a mountain.", + "input": "", + "output": "1. Majestic\n2. Rugged\n3. Towering", + "output_2": "1. Majestic\n2. Rugged\n3. Towering\n\n(Note: The words provided are adjectives and not sentences, hence they cannot be converted into passive voice as they do not contain a verb.)" + }, + { + "instruction": "Determine the state of matter for helium at room temperature.", + "input": "", + "output": "Helium is in a gaseous state at room temperature.", + "output_2": "At room temperature, helium is found in a gaseous state." + }, + { + "instruction": "Rewrite the following sentence to use a simile: The car raced down the road.", + "input": "", + "output": "The car raced down the road like a bullet.", + "output_2": "Down the road like a bullet, the car was raced." + }, + { + "instruction": "What is the symbol for the chemical element iron?", + "input": "", + "output": "The symbol for the chemical element iron is Fe.", + "output_2": "The symbol for the chemical element iron is represented by Fe." + }, + { + "instruction": "Convert the sentence into a question by adding a question tag.", + "input": "You will help me", + "output": "You will help me, won't you?", + "output_2": "I will be helped by you, won't I?" + }, + { + "instruction": "List two synonyms for \"intelligent.\"", + "input": "", + "output": "Two synonyms for \"intelligent\" are \"smart\" and \"clever.\"", + "output_2": "\"Smart\" and \"clever\" are two synonyms for \"intelligent.\"" + }, + { + "instruction": "Name the process by which plants lose water vapor through their leaves.", + "input": "", + "output": "The process by which plants lose water vapor through their leaves is called transpiration.", + "output_2": "Transpiration is called the process by which water vapor is lost through the leaves of plants." + }, + { + "instruction": "Provide a word that rhymes with \"care.\"", + "input": "", + "output": "A word that rhymes with \"care\" is \"fare.\"", + "output_2": "A word with which \"care\" rhymes is \"fare.\"" + }, + { + "instruction": "Rewrite this sentence to start with \"Despite\": She managed to finish her work, although she was tired.", + "input": "", + "output": "Despite being tired, she managed to finish her work.", + "output_2": "Despite being tired, the work was managed to be finished by her." + }, + { + "instruction": "Convert the following sentence to use a relative clause: The man is a teacher. He lives next door.", + "input": "", + "output": "The man who lives next door is a teacher.", + "output_2": "The man who is lived next door by is a teacher." + }, + { + "instruction": "Arrange the following events in chronological order: Invention of the airplane, Fall of the Berlin Wall, Discovery of America.", + "input": "", + "output": "1. Discovery of America\n2. Invention of the airplane\n3. Fall of the Berlin Wall", + "output_2": "1. America was discovered.\n2. The airplane was invented.\n3. The Berlin Wall was fallen." + }, + { + "instruction": "Name three types of clouds.", + "input": "", + "output": "Three types of clouds are cumulus, stratus, and cirrus.", + "output_2": "Cumulus, stratus, and cirrus are three types of clouds." + }, + { + "instruction": "Create a compound word using \"book.\"", + "input": "", + "output": "A compound word using \"book\" is \"bookshelf.\"", + "output_2": "A compound word using \"book\" is referred to as \"bookshelf.\"" + }, + { + "instruction": "Identify the main subject in the sentence: The noisy children played in the park.", + "input": "", + "output": "The main subject in the sentence is \"The noisy children.\"", + "output_2": "\"The noisy children\" is the main subject in the sentence." + }, + { + "instruction": "Explain what a haiku is.", + "input": "", + "output": "A haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5.", + "output_2": "A form of traditional Japanese poetry known as a haiku consists of three lines with a syllable pattern of 5-7-5." + }, + { + "instruction": "Provide the past tense form of the verb \"run.\"", + "input": "", + "output": "The past tense of the verb \"run\" is \"ran.\"", + "output_2": "The past tense of the verb \"run\" is referred to as \"ran.\"" + }, + { + "instruction": "List the ingredients required to make a basic salad.", + "input": "", + "output": "Ingredients for a basic salad typically include lettuce, tomatoes, cucumbers, and dressing.", + "output_2": "Lettuce, tomatoes, cucumbers, and dressing are typically included as ingredients for a basic salad." + }, + { + "instruction": "What is the capital of Australia?", + "input": "", + "output": "The capital of Australia is Canberra.", + "output_2": "Canberra is the capital of Australia." + }, + { + "instruction": "Name a punctuation mark used to indicate possession.", + "input": "", + "output": "A punctuation mark used to indicate possession is the apostrophe (').", + "output_2": "A punctuation mark used to indicate possession is indicated by the apostrophe (')." + }, + { + "instruction": "Name an antonym for \"joy.\"", + "input": "", + "output": "An antonym for \"joy\" is \"sorrow.\"", + "output_2": "An antonym for \"joy\" is considered to be \"sorrow.\"" + }, + { + "instruction": "Classify the following items: apple, diamond, fern.", + "input": "", + "output": "Plants: Fern\nMinerals: Diamond\nFruits: Apple", + "output_2": "Plants: Fern is grown.\nMinerals: Diamond is formed.\nFruits: Apple is cultivated." + }, + { + "instruction": "Generate three adjectives to describe the ocean.", + "input": "", + "output": "1. Vast\n2. Mysterious\n3. Serene", + "output_2": "1. Vast is remained.\n2. Mysterious is considered.\n3. Serene is felt." + }, + { + "instruction": "Determine the state of matter for nitrogen at room temperature.", + "input": "", + "output": "Nitrogen is in a gaseous state at room temperature.", + "output_2": "At room temperature, nitrogen is in a gaseous state." + }, + { + "instruction": "Convert the following statement into a question.", + "input": "You were at the concert last night.", + "output": "Were you at the concert last night?", + "output_2": "Was the concert attended by you last night?" + }, + { + "instruction": "Arrange these historical periods in chronological order: Renaissance, Roman Empire, Victorian Era.", + "input": "", + "output": "1. Roman Empire\n2. Renaissance\n3. Victorian Era", + "output_2": "1. Roman Empire\n2. Renaissance\n3. Victorian Era" + }, + { + "instruction": "Suggest a more formal synonym for \"happy.\"", + "input": "", + "output": "A more formal synonym for \"happy\" is \"content.\"", + "output_2": "A more formal synonym for \"happy\" is considered to be \"content.\"" + }, + { + "instruction": "Evaluate this sentence for spelling and grammar mistakes.", + "input": "I recieved an invitation to an exciting evenement.", + "output": "There are two errors in the sentence. The corrected sentence should be: \"I received an invitation to an exciting event.\"", + "output_2": "Two errors are found in the sentence. The corrected sentence should be: \"An invitation to an exciting event was received by me.\"" + }, + { + "instruction": "What is the capital of Germany?", + "input": "", + "output": "The capital city of Germany is Berlin.", + "output_2": "Berlin is the capital city of Germany." + }, + { + "instruction": "Classify the following into animals, plants, and metals.", + "input": "Rose, gold, cat", + "output": "Animals: Cat\nPlants: Rose\nMetals: Gold", + "output_2": "Animals: Cat is mentioned.\nPlants: Rose is mentioned.\nMetals: Gold is mentioned." + }, + { + "instruction": "Generate three verbs that mean the same as \"to run.\"", + "input": "", + "output": "1. To sprint\n2. To dash\n3. To jog", + "output_2": "1. To be sprinted\n2. To be dashed\n3. To be jogged" + }, + { + "instruction": "Look up the melting point of iron.", + "input": "", + "output": "The melting point of iron is approximately 1538 degrees Celsius or 2800 degrees Fahrenheit.", + "output_2": "The melting point of iron is approximately 1538 degrees Celsius or 2800 degrees Fahrenheit." + }, + { + "instruction": "Rewrite the following sentence using passive voice.", + "input": "The team achieved great results.", + "output": "Great results were achieved by the team.", + "output_2": "The team achieved great results." + }, + { + "instruction": "Edit the following sentence to make it more formal.", + "input": "He's got to see it to believe it.", + "output": "He must see it to believe it.", + "output_2": "It must be seen by him to be believed." + }, + { + "instruction": "Arrange the given numbers in descending order.", + "input": "7, 10, 3, 6, 9", + "output": "10, 9, 7, 6, 3.", + "output_2": "10, 9, 7, 6, 3 were counted." + }, + { + "instruction": "Given a sentence, convert it into indirect speech.", + "input": "She said, \"I am tired.\"", + "output": "She said that she was tired.", + "output_2": "It was said by her that she was tired." + }, + { + "instruction": "Transform the following sentence into the passive voice.", + "input": "They will build a new bridge.", + "output": "A new bridge will be built by them.", + "output_2": "A new bridge will be built by them." + }, + { + "instruction": "What is the capital of Italy?", + "input": "", + "output": "The capital of Italy is Rome.", + "output_2": "Rome is the capital of Italy." + }, + { + "instruction": "Generate a random alphanumeric code of 10 characters.", + "input": "", + "output": "Here is a randomly generated 10-character alphanumeric code: 3F8G6H7JQ2.", + "output_2": "A randomly generated 10-character alphanumeric code is presented here: 3F8G6H7JQ2." + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Necesary", + "output": "The correct spelling is \"Necessary\".", + "output_2": "The correct spelling is said to be \"Necessary\"." + }, + { + "instruction": "Change this sentence into the past continuous tense.", + "input": "She sings in the choir.", + "output": "She was singing in the choir.", + "output_2": "Singing in the choir was being done by her." + }, + { + "instruction": "Make the first sentence a dependent clause.", + "input": "I finished my work. I went home.", + "output": "After finishing my work, I went home.", + "output_2": "After my work was finished, I was gone home." + }, + { + "instruction": "Rewrite the sentence in future perfect tense.", + "input": "They will complete the project.", + "output": "They will have completed the project.", + "output_2": "The project will have been completed by them." + }, + { + "instruction": "Convert the given sentence into active voice.", + "input": "The law was passed by the government.", + "output": "The government passed the law.", + "output_2": "The law was passed by the government." + }, + { + "instruction": "Transform the following sentence into a question using \"could.\"", + "input": "You can help me tomorrow.", + "output": "Could you help me tomorrow?", + "output_2": "Could I be helped by you tomorrow?" + }, + { + "instruction": "Correct the grammatical errors in this sentence.", + "input": "He don't know nothing about it.", + "output": "He doesn't know anything about it.", + "output_2": "Anything about it is not known by him." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "The actual fact is that he was late.", + "output": "The fact is that he was late.", + "output_2": "The fact is that he was late." + }, + { + "instruction": "Rewrite the following sentence so that it is in active voice.", + "input": "The cookies were eaten by the children.", + "output": "The children ate the cookies.", + "output_2": "The cookies were eaten by the children." + }, + { + "instruction": "What is the contraction for \"will not\"?", + "input": "", + "output": "The contraction for \"will not\" is \"won't\".", + "output_2": "The contraction for \"will not\" is referred to as \"won't\"." + }, + { + "instruction": "Generate a sentence that follows the pattern: \"Never have I ever _____ without _____\"", + "input": "", + "output": "Never have I ever traveled without a map.", + "output_2": "Never has a map not been traveled with by me." + }, + { + "instruction": "Classify this text as a technical document or a narrative.", + "input": "This manual provides instructions for installing the software.", + "output": "Technical document", + "output_2": "A technical document" + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Zebra, Elephant, Crocodile", + "output": "sorted in alphabetical order: Crocodile, Elephant, Zebra", + "output_2": "Sorted in alphabetical order: Crocodile, Elephant, Zebra." + }, + { + "instruction": "Name the three main types of rocks.", + "input": "", + "output": "The three main types of rocks are igneous, sedimentary, and metamorphic.", + "output_2": "The three main types of rocks are identified as igneous, sedimentary, and metamorphic." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "We goes to the park every Sunday.", + "output": "We go to the park every Sunday.", + "output_2": "The park is visited by us every Sunday." + }, + { + "instruction": "Find a rhyming word for the word \"care.\"", + "input": "", + "output": "A rhyming word for the word \"care\" could be \"bare.\"", + "output_2": "A rhyming word for the word \"care\" could be \"bare\" by \"bare.\"" + }, + { + "instruction": "Reword the following sentence to the future tense.", + "input": "He is reading a novel inspired by his grandmother.", + "output": "He will be reading a novel inspired by his grandmother.", + "output_2": "A novel inspired by his grandmother will be being read by him." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Watermelon", + "output": "The fruit \"Watermelon\" is typically classified as a type of melon.", + "output_2": "The fruit \"Watermelon\" is typically classified as a type of melon by people." + }, + { + "instruction": "Suggest a continuation for the sentence: \"When the sun set,\"", + "input": "", + "output": "\"the sky turned a brilliant shade of orange, casting a warm glow over the city.\"", + "output_2": "A brilliant shade of orange was turned by the sky, casting a warm glow over the city." + }, + { + "instruction": "Classify the given sentiment as either positive, negative, or neutral.", + "input": "This is a great opportunity!", + "output": "Positive", + "output_2": "Positive feedback is appreciated." + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just returned from a vacation in Spain.", + "output": "How was your vacation in Spain?", + "output_2": "How was your vacation in Spain enjoyed by you?" + }, + { + "instruction": "Create a sentence that begins with \"If only,\"", + "input": "", + "output": "If only I had known the truth earlier...", + "output_2": "If only the truth had been known to me earlier..." + }, + { + "instruction": "Correct any spelling mistakes in the given sentence.", + "input": "I prefer homemade cookies to store boaght.", + "output": "I prefer homemade cookies to store bought.", + "output_2": "Homemade cookies are preferred by me to store-bought." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "He response to the issue was quick.", + "output": "His response to the issue was quick.", + "output_2": "A quick response was given to the issue by him." + }, + { + "instruction": "Name three essential elements found in nature.", + "input": "", + "output": "1. Carbon\n2. Oxygen\n3. Iron", + "output_2": "1. Carbon is converted.\n2. Oxygen is converted.\n3. Iron is converted." + }, + { + "instruction": "Name a dwarf planet in our solar system.", + "input": "", + "output": "One dwarf planet in our solar system is Pluto.", + "output_2": "Pluto is one dwarf planet in our solar system." + }, + { + "instruction": "Name a dwarf planet in our solar system.", + "input": "", + "output": "One dwarf planet in our solar system is Pluto.", + "output_2": "Pluto is one dwarf planet in our solar system." + }, + { + "instruction": "Classify the following words by their grammatical categories: swim, beautiful, quickly", + "input": "", + "output": "Swim: Verb\nBeautiful: Adjective\nQuickly: Adverb", + "output_2": "Swum: Verb\nBeautiful: Adjective\nQuickly: Adverb" + }, + { + "instruction": "Pick out the correct verb from the following list.", + "input": "run, mountain, quickly", + "output": "The correct verb from the list is \"run.\"", + "output_2": "The correct verb from the list, \"run,\" is chosen." + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She did not attend the meeting. She was ill.", + "output": "She did not attend the meeting because she was ill.", + "output_2": "The meeting was not attended by her because she was ill." + }, + { + "instruction": "Classify the item as either a grain or a fruit.", + "input": "Wheat", + "output": "Grain", + "output_2": "Grain is grown." + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer crashed.", + "output": "The statement \"My computer crashed\" can be classified as negative.", + "output_2": "The statement \"My computer crashed\" can be classified as negative." + }, + { + "instruction": "Rewrite the following sentence in the active voice", + "input": "The meeting will be held by the board.", + "output": "The board will hold the meeting.", + "output_2": "The meeting will be held by the board." + }, + { + "instruction": "Re-word this sentence using an active voice.", + "input": "The decision was made by the team.", + "output": "The team made the decision.", + "output_2": "The decision was made by the team." + }, + { + "instruction": "Change the given verb to its present perfect form.", + "input": "go", + "output": "Gone.", + "output_2": "Gone." + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I hate this song!", + "output": "The sentiment of the sentence \"I hate this song!\" is negative. This is indicated by the word \"hate\" which shows strong aversion or dislike towards the subject, in this case, the song.", + "output_2": "The sentiment of the sentence \"I hate this song!\" is identified as negative. This is indicated by the word \"hate,\" which shows strong aversion or dislike towards the subject, in this case, the song." + }, + { + "instruction": "Find the temperature of boiling water under normal atmospheric pressure.", + "input": "", + "output": "Under normal atmospheric pressure, the temperature of boiling water is 100 degrees Celsius (212 degrees Fahrenheit).", + "output_2": "Under normal atmospheric pressure, the boiling point of water is reached at 100 degrees Celsius (212 degrees Fahrenheit)." + }, + { + "instruction": "Edit the following sentence to correct any grammar errors.", + "input": "They was trying to fix the car all day.", + "output": "They were trying to fix the car all day.", + "output_2": "The car was being tried to be fixed all day." + }, + { + "instruction": "List 5 different types of vegetables", + "input": "", + "output": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato (biologically a fruit, but often classified as a vegetable)\n5. Spinach", + "output_2": "1. Carrot is grown.\n2. Broccoli is cultivated.\n3. Cucumber is raised.\n4. Tomato is produced (biologically a fruit, but often classified as a vegetable).\n5. Spinach is farmed." + }, + { + "instruction": "Turn the following statement into a question.", + "input": "The Golden State Warriors are a professional basketball team based in San Francisco.", + "output": "Are the Golden State Warriors a professional basketball team based in San Francisco?", + "output_2": "Is a professional basketball team based in San Francisco represented by the Golden State Warriors?" + }, + { + "instruction": "Fill in the blank to complete the sentence:\nThe most popular fruit in the world is the ____", + "input": "", + "output": "The most popular fruit in the world is the banana.", + "output_2": "The banana is the most popular fruit in the world." + }, + { + "instruction": "Generate a question based on the following statement.", + "input": "Plants produce oxygen through the process of photosynthesis.", + "output": "How do plants produce oxygen?", + "output_2": "How is oxygen produced by plants?" + }, + { + "instruction": "Correct any mistakes in the following sentence.", + "input": "Everyones going to the game tomorrow.", + "output": "Everyone's going to the game tomorrow.", + "output_2": "The game is going to be attended by everyone tomorrow." + }, + { + "instruction": "Rewrite the following sentence by adding one adverb.", + "input": "The cat slept on the mat.", + "output": "The cat peacefully slept on the mat.", + "output_2": "On the mat, the cat was peacefully slept." + }, + { + "instruction": "Identify the tense used in the sentence.", + "input": "They are watching a movie.", + "output": "The tense used in the sentence \"They are watching a movie\" is the present continuous tense.", + "output_2": "The tense used in the sentence \"A movie is being watched by them\" is the present continuous tense." + }, + { + "instruction": "Categorize the following sentence as either factual or opinion-based.", + "input": "Chocolate is the best dessert.", + "output": "Opinion-based.", + "output_2": "Opinion-based was written." + }, + { + "instruction": "Classify each sentence as either declarative, interrogative, imperative, or exclamatory.", + "input": "Can you help me?", + "output": "The classification of the sentence \"Can you help me?\" is interrogative.", + "output_2": "The sentence \"Can you help me?\" is classified as interrogative." + }, + { + "instruction": "Reverse this array", + "input": "[10, 20, 30, 40, 50]", + "output": "Here is the reversed array: [50, 40, 30, 20, 10]", + "output_2": "The reversed array is presented here: [50, 40, 30, 20, 10]." + }, + { + "instruction": "What is the chemical formula for salt?", + "input": "", + "output": "The chemical formula for salt (sodium chloride) is NaCl.", + "output_2": "The chemical formula for salt (sodium chloride) is given as NaCl." + }, + { + "instruction": "Rewrite this sentence in the present continuous tense.", + "input": "They play football every Sunday.", + "output": "They are playing football.", + "output_2": "Football is being played by them." + }, + { + "instruction": "Categorize the following items into herbs and spices.", + "input": "Cinnamon, Basil, Pepper", + "output": "Herbs: Basil\nSpices: Cinnamon, Pepper", + "output_2": "Herbs: Basil is used.\nSpices: Cinnamon, Pepper are used." + }, + { + "instruction": "Cut the following sentence into two, keeping the original meaning intact.", + "input": "She started her homework but was soon distracted by a phone call.", + "output": "She started her homework. However, she was soon distracted by a phone call.", + "output_2": "Her homework was started by her. However, she was soon distracted by a phone call." + }, + { + "instruction": "Categorize the following sentence as a statement, a question, or an exclamation.", + "input": "What a beautiful day!", + "output": "Exclamation.", + "output_2": "Exclamation was made." + }, + { + "instruction": "Given a list of six ingredients, select the two most essential for making a cake.", + "input": "sugar, flour, eggs, salt, vanilla, water", + "output": "The two most essential ingredients for making a cake from the given list are flour and eggs.", + "output_2": "The two most essential ingredients for making a cake from the given list, flour and eggs, are considered." + }, + { + "instruction": "Generate a sentence that uses the word \u2018mystic\u2019.", + "input": "", + "output": "The ancient ruins had a mystic quality that captivated all who visited.", + "output_2": "A mystic quality was had by the ancient ruins that captivated all who visited." + }, + { + "instruction": "Change the sentences from active voice to passive voice.", + "input": "The chef will prepare the meal.", + "output": "The meal will be prepared by the chef.", + "output_2": "The meal will be prepared by the chef." + }, + { + "instruction": "Transform the given statement into an imperative sentence.", + "input": "I would like you to turn off the light.", + "output": "Please turn off the light.", + "output_2": "The light is requested to be turned off." + }, + { + "instruction": "Edit the sentence to make it more formal.", + "input": "You're gonna want to see this.", + "output": "You will want to observe this.", + "output_2": "This will be wanted to be observed by you." + }, + { + "instruction": "Give me a synonym for the word 'strong'.", + "input": "", + "output": "Powerful.", + "output_2": "Powerful." + }, + { + "instruction": "Rewrite the following sentence in a more formal way:\n\"I need to find out what's wrong.\"", + "input": "", + "output": "I must ascertain what is incorrect.", + "output_2": "What is incorrect must be ascertained by me." + }, + { + "instruction": "Generate a humorous anecdote.", + "input": "", + "output": "Why was the math book sad? Because it had too many problems!", + "output_2": "Why was the math book sad? Because too many problems were had by it!" + }, + { + "instruction": "Change the sentence from passive to active voice.", + "input": "The song was sung by the choir.", + "output": "The choir sang the song.", + "output_2": "The song was sung by the choir." + }, + { + "instruction": "Label whether the following statements are true or false.", + "input": "The moon is a planet.", + "output": "False", + "output_2": "False is not a sentence and therefore cannot be converted into passive voice." + }, + { + "instruction": "Combine the given sentences into one cohesive sentence.", + "input": "It is raining. I cannot go to the park.", + "output": "Since it is raining, I cannot go to the park.", + "output_2": "Since it is raining, the park cannot be gone to by me." + }, + { + "instruction": "Rewrite the given passage in the past tense.", + "input": "They are building a new library.", + "output": "They built a new library.", + "output_2": "A new library was built." + }, + { + "instruction": "Change the following sentence so that it is in the past tense.", + "input": "He runs to the store.", + "output": "He ran to the store.", + "output_2": "The store was run to by him." + }, + { + "instruction": "Edit the following sentence to remove all passive voice.", + "input": "The song was composed by the artist.", + "output": "The artist composed the song.", + "output_2": "The song was composed by the artist." + }, + { + "instruction": "Edit the sentence below to remove any redundant words and make it more concise.", + "input": "I actually think that", + "output": "I think.", + "output_2": "It is thought by me." + }, + { + "instruction": "Edit the following sentence to remove any redundant words.", + "input": "My personal belief is that it could potentially be dangerous.", + "output": "I believe it could be dangerous.", + "output_2": "It is believed by me that it could be dangerous." + }, + { + "instruction": "Sort the following numbers in ascending order.", + "input": "5, 18, 3, 11.", + "output": "The numbers in ascending order would be: 3, 5, 11, 18.", + "output_2": "In ascending order, the numbers would be: 3, 5, 11, 18." + }, + { + "instruction": "Classify this speech as either formal or informal.", + "input": "Could you please pass the salt?", + "output": "This speech is formal.", + "output_2": "This speech is considered formal." + }, + { + "instruction": "Classify the following online review as being positive or negative.", + "input": "I found this book to be incredibly insightful and informative.", + "output": "The online review is positive.", + "output_2": "The online review is considered positive." + }, + { + "instruction": "Reverse the following list alphabetically.", + "input": "apple, orange, banana, melon", + "output": "The reverse alphabetical order of the list is: orange, melon, banana, apple.", + "output_2": "The reverse alphabetical order of the list is given as: orange, melon, banana, apple." + }, + { + "instruction": "Rearrange the following list into descending order.", + "input": "5, -3, 20, 10", + "output": "20, 10, 5, -3", + "output_2": "20, 10, 5, -3 were converted." + }, + { + "instruction": "Convert the phrase into a complete sentence.", + "input": "night sky", + "output": "The night sky was dotted with countless twinkling stars.", + "output_2": "Countless twinkling stars dotted the night sky." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Mercury, oxygen, wood", + "output": "Mercury - Liquid\nOxygen - Gas\nWood - Solid", + "output_2": "Mercury - Liquidated\nOxygen - Gassed\nWood - Solidified" + }, + { + "instruction": "Identify the correct word to complete the sentence.", + "input": "She presented her ideas with _____.", + "output": "clarity", + "output_2": "Clarity is provided." + }, + { + "instruction": "Insert a suitable adjective to complete the sentence.", + "input": "The movie was _____.", + "output": "The movie was thrilling.", + "output_2": "The audience was thrilled by the movie." + }, + { + "instruction": "Alter the content of the sentence to use the past tense.", + "input": "The ship sails at dawn.", + "output": "The ship sailed at dawn.", + "output_2": "At dawn, the ship was sailed." + }, + { + "instruction": "Re-write the following sentence to use a different verb.", + "input": "I will help you with your homework.", + "output": "I will assist you with your homework.", + "output_2": "You will be assisted with your homework by me." + }, + { + "instruction": "Pick out the odd word out in each sentence:", + "input": "Pen, Book, Tree", + "output": "The odd word out in the sentence \"Pen, Book, Tree\" is \"Tree\".", + "output_2": "The odd word out in the sentence \"Pen, Book, Tree\" is identified as \"Tree\"." + }, + { + "instruction": "Classify the following text into either fiction or non-fiction.", + "input": "The documentary covers the impact of climate change on polar bears.", + "output": "Non-fiction.", + "output_2": "Non-fiction is written." + }, + { + "instruction": "What type of number is 0?", + "input": "", + "output": "0 is an integer.", + "output_2": "An integer is represented by 0." + }, + { + "instruction": "Split the following sentence into two declarative sentences: \"The book is interesting but the cover is dull\".", + "input": "", + "output": "The book is interesting.\nThe cover is dull.", + "output_2": "The book is found interesting.\nThe cover is considered dull." + }, + { + "instruction": "Classify this sentence into one of the following topics: health, economy, technology, music", + "input": "Spotify has revolutionized how we listen to music.", + "output": "The topic of the given sentence is music.", + "output_2": "The topic of music is given in the sentence." + }, + { + "instruction": "Generate a question from the following statement", + "input": "Leonardo da Vinci painted the Mona Lisa.", + "output": "Who painted the Mona Lisa?", + "output_2": "By whom was the Mona Lisa painted?" + }, + { + "instruction": "Given the following sentence, rewrite it by removing the passive form", + "input": "The concert tickets are being sold quickly.", + "output": "They are selling the concert tickets quickly.", + "output_2": "The concert tickets are being sold quickly." + }, + { + "instruction": "Given a sentence, count the number of words.", + "input": "The quick brown fox jumps over the lazy dog.", + "output": "The sentence \"The quick brown fox jumps over the lazy dog.\" contains 9 words.", + "output_2": "It is contained by the sentence \"The quick brown fox jumps over the lazy dog.\" that there are 9 words." + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The meal was not delicious.", + "output": "The meal could use some improvement.", + "output_2": "Some improvement could be used in the meal." + }, + { + "instruction": "Rewrite the sentence so that it does not contain any passive voice.", + "input": "The song was played on the radio.", + "output": "The radio played the song.", + "output_2": "The song was played by the radio." + }, + { + "instruction": "Find a synonym for the word 'start'.", + "input": "", + "output": "A synonym for the word 'start' could be \"commence.\"", + "output_2": "A synonym for the word 'start' could be \"commence\" by some." + }, + { + "instruction": "Change the sentence \"This should not have happened\" into a declarative sentence.", + "input": "", + "output": "This should have happened differently.", + "output_2": "Differently, this should have happened." + }, + { + "instruction": "Rewrite the sentence: \"The poem was composed in a lyrical manner.\"", + "input": "", + "output": "The poem was crafted lyrically.", + "output_2": "The poem was crafted lyrically." + }, + { + "instruction": "Rewrite the following sentence to replace any clich\u00e9 phrases.", + "input": "It's a piece of cake", + "output": "It's very easy.", + "output_2": "It is considered very easy." + }, + { + "instruction": "Pick the best synonym for the given word.", + "input": "Gather", + "output": "Collect.", + "output_2": "Be collected." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "The storm started suddenly.", + "output": "The storm came out of the blue.", + "output_2": "The storm was come out of the blue." + }, + { + "instruction": "Scramble the given phrase.", + "input": "A random phrase", + "output": "A nmaodr hrasep", + "output_2": "A nmaodr hrasep was written." + }, + { + "instruction": "rewrite this sentence as a question.", + "input": "Your brother went to London last year.", + "output": "Did your brother go to London last year?", + "output_2": "Was London visited by your brother last year?" + }, + { + "instruction": "Rewrite the sentence below so that it is grammatically correct.", + "input": "Him and I seen the movie.", + "output": "He and I saw the movie.", + "output_2": "The movie was seen by him and me." + }, + { + "instruction": "Edit the given text to ensure all plural nouns are spelled correctly.", + "input": "The boxes was stacked neatly.", + "output": "The boxes were stacked neatly.", + "output_2": "The boxes had been stacked neatly." + }, + { + "instruction": "Rewrite this English sentence into passive voice.", + "input": "Alex fixed the car.", + "output": "The car was fixed by Alex.", + "output_2": "The car was fixed by Alex." + }, + { + "instruction": "Rewrite this sentence in the form of a question.", + "input": "This is the biggest tree in the park.", + "output": "Is this the biggest tree in the park?", + "output_2": "Is this tree considered the biggest in the park?" + }, + { + "instruction": "Edit the text so that it is grammatically correct.", + "input": "He don't like going there", + "output": "He doesn't like going there.", + "output_2": "Going there is not liked by him." + }, + { + "instruction": "Identify the hypernym for the word \"car\".", + "input": "", + "output": "The hypernym for the word \"car\" is \"vehicle\".", + "output_2": "The hypernym for the word \"vehicle\" is \"car\"." + }, + { + "instruction": "Construct a sentence using the words \"achieve\", \"goal\", and \"perseverance\".", + "input": "", + "output": "With perseverance, you can achieve any goal.", + "output_2": "With perseverance, any goal can be achieved by you." + }, + { + "instruction": "Rewrite this statement as an imperative sentence.", + "input": "You should check the document for errors.", + "output": "Check the document for errors.", + "output_2": "The document should be checked for errors." + }, + { + "instruction": "Capitalize each word in the sentence.", + "input": "i love watching the sunrise", + "output": "I Love Watching The Sunrise", + "output_2": "The sunrise is loved to be watched by me." + }, + { + "instruction": "Create a new sentence by combining the two sentences.", + "input": "He likes to swim. He goes to the pool every day.", + "output": "He likes to swim and goes to the pool every day.", + "output_2": "Swimming is liked by him, and the pool is gone to by him every day." + }, + { + "instruction": "Combine the two ideas in the sentence.", + "input": "Learning a new language is useful. It can be challenging.", + "output": "Learning a new language is useful but can be challenging.", + "output_2": "A new language being learned is useful but can be challenging." + }, + { + "instruction": "Reword the sentence to use other words without changing the meaning.", + "input": "It rained heavily during the night.", + "output": "The night experienced significant rainfall.", + "output_2": "Significant rainfall was experienced by the night." + }, + { + "instruction": "Rewrite the sentence to use a negative adverb.", + "input": "She always remembers to call.", + "output": "She never forgets to call.", + "output_2": "It is never forgotten by her to call." + }, + { + "instruction": "Categorize the given list of animals.", + "input": "Shark, Dolphin, Trout", + "output": "Fish: Shark, Trout\nMammals: Dolphin", + "output_2": "Fish: Shark, Trout are listed.\nMammals: Dolphin is listed." + }, + { + "instruction": "Rewrite the sentence using a synonym.", + "input": "It's cold outside.", + "output": "It's chilly outside.", + "output_2": "It is considered chilly outside." + }, + { + "instruction": "Change the given sentence into passive voice.", + "input": "They celebrate the festival annually.", + "output": "The festival is celebrated annually.", + "output_2": "The festival is celebrated annually." + }, + { + "instruction": "Generate a question with the keyword \u201cbacteria\u201d", + "input": "", + "output": "How do bacteria affect human health?", + "output_2": "How is human health affected by bacteria?" + }, + { + "instruction": "Take the following noun and turn it into a verb.", + "input": "Light", + "output": "To light.", + "output_2": "To be lit." + }, + { + "instruction": "Evaluate the following phrase by transforming it into the spelling given.", + "input": "freind --> friend", + "output": "The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\".", + "output_2": "The spelling of the given phrase \"freind\" is identified as incorrect; the correct spelling is identified as \"friend\"." + }, + { + "instruction": "Select the noun phrase in the following sentence.", + "input": "He enjoyed a delicious cake at the party.", + "output": "The noun phrases in the given sentence are \"a delicious cake\" and \"the party.\"", + "output_2": "The noun phrases \"a delicious cake\" and \"the party\" are identified in the given sentence." + }, + { + "instruction": "Rewrite the following phrase in the present tense.", + "input": "They had played soccer all afternoon.", + "output": "They play soccer all afternoon.", + "output_2": "Soccer is played by them all afternoon." + }, + { + "instruction": "Rewrite the following phrase as a question.", + "input": "We must finish the project by tomorrow.", + "output": "Must we finish the project by tomorrow?", + "output_2": "Must the project be finished by us by tomorrow?" + }, + { + "instruction": "Transform the sentence into a negative statement.", + "input": "She always arrives on time.", + "output": "She never arrives late.", + "output_2": "Late is never arrived at by her." + }, + { + "instruction": "Suggest a simpler word to replace the word in braces.", + "input": "He moved {expeditiously} through the crowd.", + "output": "He moved quickly through the crowd.", + "output_2": "Quickly, the crowd was moved through by him." + }, + { + "instruction": "Generate a five-word phrase using the words \"moon,\" \"crisp,\" and \"night.\"", + "input": "", + "output": "\"Crisp moon illuminates the night.\"", + "output_2": "The night is illuminated by the crisp moon." + }, + { + "instruction": "Describe the color blue in 3 words.", + "input": "", + "output": "Deep, calming, cool.", + "output_2": "Deeply calmed, cooled." + }, + { + "instruction": "Classify the following species as dogs or cats.", + "input": "Siamese", + "output": "Siamese is a cat.", + "output_2": "A cat is Siamese." + }, + { + "instruction": "Classify an input string as either a noun or a verb.", + "input": "jump", + "output": "\"jump\" can be classified as a verb.", + "output_2": "\"Jump\" can be classified as a verb." + }, + { + "instruction": "Insert the correct punctuation at the provided places in the following sentence.", + "input": "Where are you going", + "output": "Where are you going?", + "output_2": "Where is it being gone to by you?" + }, + { + "instruction": "Convert 10 feet to meters.", + "input": "", + "output": "10 feet is equal to approximately 3.048 meters.", + "output_2": "Approximately 3.048 meters is equal to 10 feet." + }, + { + "instruction": "Simplify the given sentence.", + "input": "The teacher explained that the reason we were studying this was because it is important.", + "output": "The teacher explained that we were studying this because it's important.", + "output_2": "It was explained by the teacher that this was being studied by us because it's important." + }, + { + "instruction": "Determine whether the given two words are synonyms or antonyms.", + "input": "Open - Close", + "output": "Open and Close are antonyms.", + "output_2": "It is acknowledged that Open and Close are antonyms." + }, + { + "instruction": "Change this sentence grammatically.", + "input": "They has seen the movie.", + "output": "They have seen the movie.", + "output_2": "The movie has been seen by them." + }, + { + "instruction": "Generate a past-tense verb that describes a person singing.", + "input": "", + "output": "Sang.", + "output_2": "Was sung." + }, + { + "instruction": "Find the synonym for the given verb.", + "input": "Begin", + "output": "Commence", + "output_2": "The following text is to be converted to passive voice: Commence." + }, + { + "instruction": "Name 3 different animals that are active during the day.", + "input": "", + "output": "1. Squirrel\n2. Eagle\n3. Tiger", + "output_2": "1. Squirrel is chased.\n2. Eagle is seen.\n3. Tiger is admired." + }, + { + "instruction": "Fill in the blanks with appropriate words.", + "input": "The sky was filled with colorful ________ at sunset.", + "output": "The sky was filled with colorful clouds at sunset.", + "output_2": "Colorful clouds filled the sky at sunset." + }, + { + "instruction": "Categorize the following items into one of the following categories: appliance, clothing, or food.", + "input": "Microwave, T-shirt, Bread", + "output": "Appliance: Microwave\nClothing: T-shirt\nFood: Bread", + "output_2": "Appliance: Microwave is used.\nClothing: T-shirt is worn.\nFood: Bread is eaten." + }, + { + "instruction": "Edit the sentence: \"We enjoys watching movies\"", + "input": "", + "output": "We enjoy watching movies.", + "output_2": "Movies are enjoyed by us." + }, + { + "instruction": "Edit the given sentence to make it grammatically correct.", + "input": "She dont have any money", + "output": "She doesn't have any money.", + "output_2": "Any money isn't had by her." + }, + { + "instruction": "Classify the following animal as carnivore, herbivore, or omnivore.", + "input": "Deer", + "output": "Herbivore", + "output_2": "I'm sorry, but the word \"Herbivore\" does not form a sentence and thus cannot be converted into passive voice. Passive voice requires a sentence with a clear subject, verb, and object, which is not applicable to a single noun." + }, + { + "instruction": "Fix the following grammar mistake.", + "input": "He want to go to the movies, but he has no money.", + "output": "He wants to go to the movies, but he has no money.", + "output_2": "It is wanted by him to go to the movies, but no money is had by him." + }, + { + "instruction": "Automatically correct the given sentence.", + "input": "I want to go to london.", + "output": "I want to go to London.", + "output_2": "London is wanted to be gone to by me." + }, + { + "instruction": "Generate a random sequence of 10 digits.", + "input": "", + "output": "Here is a randomly generated sequence of 10 digits: 9274036185.", + "output_2": "A randomly generated sequence of 10 digits, 9274036185, is presented here." + }, + { + "instruction": "Generate an antonym for the word 'optimistic'.", + "input": "", + "output": "An antonym for the word 'optimistic' could be 'pessimistic'.", + "output_2": "An antonym for the word 'optimistic' could be considered 'pessimistic'." + }, + { + "instruction": "Edit the text for grammar and punctuation.", + "input": "today was amazing I went hiking and saw a bear", + "output": "\"Today was amazing. I went hiking and saw a bear.\"", + "output_2": "Today was amazing. Hiking was gone on by me and a bear was seen." + }, + { + "instruction": "Rewrite this sentence to eliminate the passive voice.", + "input": "The cake was made by Susan.", + "output": "Susan made the cake.", + "output_2": "The cake was made by Susan." + }, + { + "instruction": "Sort the following array in ascending order.", + "input": "[9, 2, 7, 4]", + "output": "Here is the sorted array in ascending order: [2, 4, 7, 9].", + "output_2": "The sorted array in ascending order is presented here: [2, 4, 7, 9]." + }, + { + "instruction": "Add the correct verb to the sentence:\nThe students _________ the quiz answers.", + "input": "", + "output": "The students discussed the quiz answers.", + "output_2": "The quiz answers were discussed by the students." + }, + { + "instruction": "Put the following words in alphabetical order: kitten, zebra, alligator.", + "input": "", + "output": "Alligator, kitten, zebra.", + "output_2": "Alligator, kitten, zebra." + }, + { + "instruction": "Delete the redundant words in the following sentence:\nWe went to see to the mountains.", + "input": "", + "output": "We went to see the mountains.", + "output_2": "The mountains were gone to see by us." + }, + { + "instruction": "Rewrite this sentence using appropriate grammar rules:", + "input": "he saw a old man", + "output": "He saw an old man.", + "output_2": "An old man was seen by him." + }, + { + "instruction": "Find the next 3 terms in the following sequence: 3, 5, 7, 9", + "input": "", + "output": "The next three terms in the sequence are: 11, 13, 15.", + "output_2": "The next three terms in the sequence are identified as: 11, 13, 15." + }, + { + "instruction": "Reverse the order of the given phrase.", + "input": "sun and moon", + "output": "moon and sun", + "output_2": "The moon and the sun are referred to." + } +] \ No newline at end of file diff --git a/Codes/ch07/02_dataset-utilities/instruction-examples.json b/Codes/ch07/02_dataset-utilities/instruction-examples.json new file mode 100644 index 0000000..d2a1570 --- /dev/null +++ b/Codes/ch07/02_dataset-utilities/instruction-examples.json @@ -0,0 +1,1002 @@ +[ + { + "instruction": "Identify the verb in the following sentence: The cat sleeps on the couch.", + "input": "", + "output": "The verb in the sentence is \"sleeps.\"" + }, + { + "instruction": "What is the plural form of \"goose\"?", + "input": "", + "output": "The plural form of \"goose\" is \"geese.\"" + }, + { + "instruction": "List three primary colors.", + "input": "", + "output": "The three primary colors are red, blue, and yellow." + }, + { + "instruction": "Change the following sentence to past perfect tense.", + "input": "They finish the game.", + "output": "They had finished the game." + }, + { + "instruction": "What is the abbreviation for \"Doctor of Philosophy\"?", + "input": "", + "output": "The abbreviation for \"Doctor of Philosophy\" is Ph.D." + }, + { + "instruction": "Correct the misspelled word in the following sentence: I recieved your message.", + "input": "", + "output": "The misspelled word \"recieved\" should be spelled \"received.\"" + }, + { + "instruction": "Transform the following sentence to include an exclamation mark to show excitement.", + "input": "I won the lottery", + "output": "I won the lottery!" + }, + { + "instruction": "What is the antonym of \"ancient\"?", + "input": "", + "output": "The antonym of \"ancient\" is \"modern.\"" + }, + { + "instruction": "Name a synonym for \"happiness.\"", + "input": "", + "output": "A synonym for \"happiness\" is \"joy.\"" + }, + { + "instruction": "Classify the following items: gold ring, daisy, python.", + "input": "", + "output": "Animals: Python\nPlants: Daisy\nMinerals: Gold ring" + }, + { + "instruction": "Generate three adjectives to describe a mountain.", + "input": "", + "output": "1. Majestic\n2. Rugged\n3. Towering" + }, + { + "instruction": "Determine the state of matter for helium at room temperature.", + "input": "", + "output": "Helium is in a gaseous state at room temperature." + }, + { + "instruction": "Rewrite the following sentence to use a simile: The car raced down the road.", + "input": "", + "output": "The car raced down the road like a bullet." + }, + { + "instruction": "What is the symbol for the chemical element iron?", + "input": "", + "output": "The symbol for the chemical element iron is Fe." + }, + { + "instruction": "Convert the sentence into a question by adding a question tag.", + "input": "You will help me", + "output": "You will help me, won't you?" + }, + { + "instruction": "List two synonyms for \"intelligent.\"", + "input": "", + "output": "Two synonyms for \"intelligent\" are \"smart\" and \"clever.\"" + }, + { + "instruction": "Name the process by which plants lose water vapor through their leaves.", + "input": "", + "output": "The process by which plants lose water vapor through their leaves is called transpiration." + }, + { + "instruction": "Provide a word that rhymes with \"care.\"", + "input": "", + "output": "A word that rhymes with \"care\" is \"fare.\"" + }, + { + "instruction": "Rewrite this sentence to start with \"Despite\": She managed to finish her work, although she was tired.", + "input": "", + "output": "Despite being tired, she managed to finish her work." + }, + { + "instruction": "Convert the following sentence to use a relative clause: The man is a teacher. He lives next door.", + "input": "", + "output": "The man who lives next door is a teacher." + }, + { + "instruction": "Arrange the following events in chronological order: Invention of the airplane, Fall of the Berlin Wall, Discovery of America.", + "input": "", + "output": "1. Discovery of America\n2. Invention of the airplane\n3. Fall of the Berlin Wall" + }, + { + "instruction": "Name three types of clouds.", + "input": "", + "output": "Three types of clouds are cumulus, stratus, and cirrus." + }, + { + "instruction": "Create a compound word using \"book.\"", + "input": "", + "output": "A compound word using \"book\" is \"bookshelf.\"" + }, + { + "instruction": "Identify the main subject in the sentence: The noisy children played in the park.", + "input": "", + "output": "The main subject in the sentence is \"The noisy children.\"" + }, + { + "instruction": "Explain what a haiku is.", + "input": "", + "output": "A haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5." + }, + { + "instruction": "Provide the past tense form of the verb \"run.\"", + "input": "", + "output": "The past tense of the verb \"run\" is \"ran.\"" + }, + { + "instruction": "List the ingredients required to make a basic salad.", + "input": "", + "output": "Ingredients for a basic salad typically include lettuce, tomatoes, cucumbers, and dressing." + }, + { + "instruction": "What is the capital of Australia?", + "input": "", + "output": "The capital of Australia is Canberra." + }, + { + "instruction": "Name a punctuation mark used to indicate possession.", + "input": "", + "output": "A punctuation mark used to indicate possession is the apostrophe (')." + }, + { + "instruction": "Name an antonym for \"joy.\"", + "input": "", + "output": "An antonym for \"joy\" is \"sorrow.\"" + }, + { + "instruction": "Classify the following items: apple, diamond, fern.", + "input": "", + "output": "Plants: Fern\nMinerals: Diamond\nFruits: Apple" + }, + { + "instruction": "Generate three adjectives to describe the ocean.", + "input": "", + "output": "1. Vast\n2. Mysterious\n3. Serene" + }, + { + "instruction": "Determine the state of matter for nitrogen at room temperature.", + "input": "", + "output": "Nitrogen is in a gaseous state at room temperature." + }, + { + "instruction": "Convert the following statement into a question.", + "input": "You were at the concert last night.", + "output": "Were you at the concert last night?" + }, + { + "instruction": "Arrange these historical periods in chronological order: Renaissance, Roman Empire, Victorian Era.", + "input": "", + "output": "1. Roman Empire\n2. Renaissance\n3. Victorian Era" + }, + { + "instruction": "Suggest a more formal synonym for \"happy.\"", + "input": "", + "output": "A more formal synonym for \"happy\" is \"content.\"" + }, + { + "instruction": "Evaluate this sentence for spelling and grammar mistakes.", + "input": "I recieved an invitation to an exciting evenement.", + "output": "There are two errors in the sentence. The corrected sentence should be: \"I received an invitation to an exciting event.\"" + }, + { + "instruction": "What is the capital of Germany?", + "input": "", + "output": "The capital city of Germany is Berlin." + }, + { + "instruction": "Classify the following into animals, plants, and metals.", + "input": "Rose, gold, cat", + "output": "Animals: Cat\nPlants: Rose\nMetals: Gold" + }, + { + "instruction": "Generate three verbs that mean the same as \"to run.\"", + "input": "", + "output": "1. To sprint\n2. To dash\n3. To jog" + }, + { + "instruction": "Look up the melting point of iron.", + "input": "", + "output": "The melting point of iron is approximately 1538 degrees Celsius or 2800 degrees Fahrenheit." + }, + { + "instruction": "Rewrite the following sentence using passive voice.", + "input": "The team achieved great results.", + "output": "Great results were achieved by the team." + }, + { + "instruction": "Edit the following sentence to make it more formal.", + "input": "He's got to see it to believe it.", + "output": "He must see it to believe it." + }, + { + "instruction": "Arrange the given numbers in descending order.", + "input": "7, 10, 3, 6, 9", + "output": "10, 9, 7, 6, 3." + }, + { + "instruction": "Given a sentence, convert it into indirect speech.", + "input": "She said, \"I am tired.\"", + "output": "She said that she was tired." + }, + { + "instruction": "Transform the following sentence into the passive voice.", + "input": "They will build a new bridge.", + "output": "A new bridge will be built by them." + }, + { + "instruction": "What is the capital of Italy?", + "input": "", + "output": "The capital of Italy is Rome." + }, + { + "instruction": "Generate a random alphanumeric code of 10 characters.", + "input": "", + "output": "Here is a randomly generated 10-character alphanumeric code: 3F8G6H7JQ2." + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Necesary", + "output": "The correct spelling is \"Necessary\"." + }, + { + "instruction": "Change this sentence into the past continuous tense.", + "input": "She sings in the choir.", + "output": "She was singing in the choir." + }, + { + "instruction": "Make the first sentence a dependent clause.", + "input": "I finished my work. I went home.", + "output": "After finishing my work, I went home." + }, + { + "instruction": "Rewrite the sentence in future perfect tense.", + "input": "They will complete the project.", + "output": "They will have completed the project." + }, + { + "instruction": "Convert the given sentence into active voice.", + "input": "The law was passed by the government.", + "output": "The government passed the law." + }, + { + "instruction": "Transform the following sentence into a question using \"could.\"", + "input": "You can help me tomorrow.", + "output": "Could you help me tomorrow?" + }, + { + "instruction": "Correct the grammatical errors in this sentence.", + "input": "He don't know nothing about it.", + "output": "He doesn't know anything about it." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "The actual fact is that he was late.", + "output": "The fact is that he was late." + }, + { + "instruction": "Rewrite the following sentence so that it is in active voice.", + "input": "The cookies were eaten by the children.", + "output": "The children ate the cookies." + }, + { + "instruction": "What is the contraction for \"will not\"?", + "input": "", + "output": "The contraction for \"will not\" is \"won't\"." + }, + { + "instruction": "Generate a sentence that follows the pattern: \"Never have I ever _____ without _____\"", + "input": "", + "output": "Never have I ever traveled without a map." + }, + { + "instruction": "Classify this text as a technical document or a narrative.", + "input": "This manual provides instructions for installing the software.", + "output": "Technical document" + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Zebra, Elephant, Crocodile", + "output": "sorted in alphabetical order: Crocodile, Elephant, Zebra" + }, + { + "instruction": "Name the three main types of rocks.", + "input": "", + "output": "The three main types of rocks are igneous, sedimentary, and metamorphic." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "We goes to the park every Sunday.", + "output": "We go to the park every Sunday." + }, + { + "instruction": "Find a rhyming word for the word \"care.\"", + "input": "", + "output": "A rhyming word for the word \"care\" could be \"bare.\"" + }, + { + "instruction": "Reword the following sentence to the future tense.", + "input": "He is reading a novel inspired by his grandmother.", + "output": "He will be reading a novel inspired by his grandmother." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Watermelon", + "output": "The fruit \"Watermelon\" is typically classified as a type of melon." + }, + { + "instruction": "Suggest a continuation for the sentence: \"When the sun set,\"", + "input": "", + "output": "\"the sky turned a brilliant shade of orange, casting a warm glow over the city.\"" + }, + { + "instruction": "Classify the given sentiment as either positive, negative, or neutral.", + "input": "This is a great opportunity!", + "output": "Positive" + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just returned from a vacation in Spain.", + "output": "How was your vacation in Spain?" + }, + { + "instruction": "Create a sentence that begins with \"If only,\"", + "input": "", + "output": "If only I had known the truth earlier..." + }, + { + "instruction": "Correct any spelling mistakes in the given sentence.", + "input": "I prefer homemade cookies to store boaght.", + "output": "I prefer homemade cookies to store bought." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "He response to the issue was quick.", + "output": "His response to the issue was quick." + }, + { + "instruction": "Name three essential elements found in nature.", + "input": "", + "output": "1. Carbon\n2. Oxygen\n3. Iron" + }, + { + "instruction": "Name a dwarf planet in our solar system.", + "input": "", + "output": "One dwarf planet in our solar system is Pluto." + }, + { + "instruction": "Name a dwarf planet in our solar system.", + "input": "", + "output": "One dwarf planet in our solar system is Pluto." + }, + { + "instruction": "Classify the following words by their grammatical categories: swim, beautiful, quickly", + "input": "", + "output": "Swim: Verb\nBeautiful: Adjective\nQuickly: Adverb" + }, + { + "instruction": "Pick out the correct verb from the following list.", + "input": "run, mountain, quickly", + "output": "The correct verb from the list is \"run.\"" + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She did not attend the meeting. She was ill.", + "output": "She did not attend the meeting because she was ill." + }, + { + "instruction": "Classify the item as either a grain or a fruit.", + "input": "Wheat", + "output": "Grain" + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer crashed.", + "output": "The statement \"My computer crashed\" can be classified as negative." + }, + { + "instruction": "Rewrite the following sentence in the active voice", + "input": "The meeting will be held by the board.", + "output": "The board will hold the meeting." + }, + { + "instruction": "Re-word this sentence using an active voice.", + "input": "The decision was made by the team.", + "output": "The team made the decision." + }, + { + "instruction": "Change the given verb to its present perfect form.", + "input": "go", + "output": "Gone." + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I hate this song!", + "output": "The sentiment of the sentence \"I hate this song!\" is negative. This is indicated by the word \"hate\" which shows strong aversion or dislike towards the subject, in this case, the song." + }, + { + "instruction": "Find the temperature of boiling water under normal atmospheric pressure.", + "input": "", + "output": "Under normal atmospheric pressure, the temperature of boiling water is 100 degrees Celsius (212 degrees Fahrenheit)." + }, + { + "instruction": "Edit the following sentence to correct any grammar errors.", + "input": "They was trying to fix the car all day.", + "output": "They were trying to fix the car all day." + }, + { + "instruction": "List 5 different types of vegetables", + "input": "", + "output": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato (biologically a fruit, but often classified as a vegetable)\n5. Spinach" + }, + { + "instruction": "Turn the following statement into a question.", + "input": "The Golden State Warriors are a professional basketball team based in San Francisco.", + "output": "Are the Golden State Warriors a professional basketball team based in San Francisco?" + }, + { + "instruction": "Fill in the blank to complete the sentence:\nThe most popular fruit in the world is the ____", + "input": "", + "output": "The most popular fruit in the world is the banana." + }, + { + "instruction": "Generate a question based on the following statement.", + "input": "Plants produce oxygen through the process of photosynthesis.", + "output": "How do plants produce oxygen?" + }, + { + "instruction": "Correct any mistakes in the following sentence.", + "input": "Everyones going to the game tomorrow.", + "output": "Everyone's going to the game tomorrow." + }, + { + "instruction": "Rewrite the following sentence by adding one adverb.", + "input": "The cat slept on the mat.", + "output": "The cat peacefully slept on the mat." + }, + { + "instruction": "Identify the tense used in the sentence.", + "input": "They are watching a movie.", + "output": "The tense used in the sentence \"They are watching a movie\" is the present continuous tense." + }, + { + "instruction": "Categorize the following sentence as either factual or opinion-based.", + "input": "Chocolate is the best dessert.", + "output": "Opinion-based." + }, + { + "instruction": "Classify each sentence as either declarative, interrogative, imperative, or exclamatory.", + "input": "Can you help me?", + "output": "The classification of the sentence \"Can you help me?\" is interrogative." + }, + { + "instruction": "Reverse this array", + "input": "[10, 20, 30, 40, 50]", + "output": "Here is the reversed array: [50, 40, 30, 20, 10]" + }, + { + "instruction": "What is the chemical formula for salt?", + "input": "", + "output": "The chemical formula for salt (sodium chloride) is NaCl." + }, + { + "instruction": "Rewrite this sentence in the present continuous tense.", + "input": "They play football every Sunday.", + "output": "They are playing football." + }, + { + "instruction": "Categorize the following items into herbs and spices.", + "input": "Cinnamon, Basil, Pepper", + "output": "Herbs: Basil\nSpices: Cinnamon, Pepper" + }, + { + "instruction": "Cut the following sentence into two, keeping the original meaning intact.", + "input": "She started her homework but was soon distracted by a phone call.", + "output": "She started her homework. However, she was soon distracted by a phone call." + }, + { + "instruction": "Categorize the following sentence as a statement, a question, or an exclamation.", + "input": "What a beautiful day!", + "output": "Exclamation." + }, + { + "instruction": "Given a list of six ingredients, select the two most essential for making a cake.", + "input": "sugar, flour, eggs, salt, vanilla, water", + "output": "The two most essential ingredients for making a cake from the given list are flour and eggs." + }, + { + "instruction": "Generate a sentence that uses the word ‘mystic’.", + "input": "", + "output": "The ancient ruins had a mystic quality that captivated all who visited." + }, + { + "instruction": "Change the sentences from active voice to passive voice.", + "input": "The chef will prepare the meal.", + "output": "The meal will be prepared by the chef." + }, + { + "instruction": "Transform the given statement into an imperative sentence.", + "input": "I would like you to turn off the light.", + "output": "Please turn off the light." + }, + { + "instruction": "Edit the sentence to make it more formal.", + "input": "You're gonna want to see this.", + "output": "You will want to observe this." + }, + { + "instruction": "Give me a synonym for the word 'strong'.", + "input": "", + "output": "Powerful." + }, + { + "instruction": "Rewrite the following sentence in a more formal way:\n\"I need to find out what's wrong.\"", + "input": "", + "output": "I must ascertain what is incorrect." + }, + { + "instruction": "Generate a humorous anecdote.", + "input": "", + "output": "Why was the math book sad? Because it had too many problems!" + }, + { + "instruction": "Change the sentence from passive to active voice.", + "input": "The song was sung by the choir.", + "output": "The choir sang the song." + }, + { + "instruction": "Label whether the following statements are true or false.", + "input": "The moon is a planet.", + "output": "False" + }, + { + "instruction": "Combine the given sentences into one cohesive sentence.", + "input": "It is raining. I cannot go to the park.", + "output": "Since it is raining, I cannot go to the park." + }, + { + "instruction": "Rewrite the given passage in the past tense.", + "input": "They are building a new library.", + "output": "They built a new library." + }, + { + "instruction": "Change the following sentence so that it is in the past tense.", + "input": "He runs to the store.", + "output": "He ran to the store." + }, + { + "instruction": "Edit the following sentence to remove all passive voice.", + "input": "The song was composed by the artist.", + "output": "The artist composed the song." + }, + { + "instruction": "Edit the sentence below to remove any redundant words and make it more concise.", + "input": "I actually think that", + "output": "I think." + }, + { + "instruction": "Edit the following sentence to remove any redundant words.", + "input": "My personal belief is that it could potentially be dangerous.", + "output": "I believe it could be dangerous." + }, + { + "instruction": "Sort the following numbers in ascending order.", + "input": "5, 18, 3, 11.", + "output": "The numbers in ascending order would be: 3, 5, 11, 18." + }, + { + "instruction": "Classify this speech as either formal or informal.", + "input": "Could you please pass the salt?", + "output": "This speech is formal." + }, + { + "instruction": "Classify the following online review as being positive or negative.", + "input": "I found this book to be incredibly insightful and informative.", + "output": "The online review is positive." + }, + { + "instruction": "Reverse the following list alphabetically.", + "input": "apple, orange, banana, melon", + "output": "The reverse alphabetical order of the list is: orange, melon, banana, apple." + }, + { + "instruction": "Rearrange the following list into descending order.", + "input": "5, -3, 20, 10", + "output": "20, 10, 5, -3" + }, + { + "instruction": "Convert the phrase into a complete sentence.", + "input": "night sky", + "output": "The night sky was dotted with countless twinkling stars." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Mercury, oxygen, wood", + "output": "Mercury - Liquid\nOxygen - Gas\nWood - Solid" + }, + { + "instruction": "Identify the correct word to complete the sentence.", + "input": "She presented her ideas with _____.", + "output": "clarity" + }, + { + "instruction": "Insert a suitable adjective to complete the sentence.", + "input": "The movie was _____.", + "output": "The movie was thrilling." + }, + { + "instruction": "Alter the content of the sentence to use the past tense.", + "input": "The ship sails at dawn.", + "output": "The ship sailed at dawn." + }, + { + "instruction": "Re-write the following sentence to use a different verb.", + "input": "I will help you with your homework.", + "output": "I will assist you with your homework." + }, + { + "instruction": "Pick out the odd word out in each sentence:", + "input": "Pen, Book, Tree", + "output": "The odd word out in the sentence \"Pen, Book, Tree\" is \"Tree\"." + }, + { + "instruction": "Classify the following text into either fiction or non-fiction.", + "input": "The documentary covers the impact of climate change on polar bears.", + "output": "Non-fiction." + }, + { + "instruction": "What type of number is 0?", + "input": "", + "output": "0 is an integer." + }, + { + "instruction": "Split the following sentence into two declarative sentences: \"The book is interesting but the cover is dull\".", + "input": "", + "output": "The book is interesting.\nThe cover is dull." + }, + { + "instruction": "Classify this sentence into one of the following topics: health, economy, technology, music", + "input": "Spotify has revolutionized how we listen to music.", + "output": "The topic of the given sentence is music." + }, + { + "instruction": "Generate a question from the following statement", + "input": "Leonardo da Vinci painted the Mona Lisa.", + "output": "Who painted the Mona Lisa?" + }, + { + "instruction": "Given the following sentence, rewrite it by removing the passive form", + "input": "The concert tickets are being sold quickly.", + "output": "They are selling the concert tickets quickly." + }, + { + "instruction": "Given a sentence, count the number of words.", + "input": "The quick brown fox jumps over the lazy dog.", + "output": "The sentence \"The quick brown fox jumps over the lazy dog.\" contains 9 words." + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The meal was not delicious.", + "output": "The meal could use some improvement." + }, + { + "instruction": "Rewrite the sentence so that it does not contain any passive voice.", + "input": "The song was played on the radio.", + "output": "The radio played the song." + }, + { + "instruction": "Find a synonym for the word 'start'.", + "input": "", + "output": "A synonym for the word 'start' could be \"commence.\"" + }, + { + "instruction": "Change the sentence \"This should not have happened\" into a declarative sentence.", + "input": "", + "output": "This should have happened differently." + }, + { + "instruction": "Rewrite the sentence: \"The poem was composed in a lyrical manner.\"", + "input": "", + "output": "The poem was crafted lyrically." + }, + { + "instruction": "Rewrite the following sentence to replace any cliché phrases.", + "input": "It's a piece of cake", + "output": "It's very easy." + }, + { + "instruction": "Pick the best synonym for the given word.", + "input": "Gather", + "output": "Collect." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "The storm started suddenly.", + "output": "The storm came out of the blue." + }, + { + "instruction": "Scramble the given phrase.", + "input": "A random phrase", + "output": "A nmaodr hrasep" + }, + { + "instruction": "rewrite this sentence as a question.", + "input": "Your brother went to London last year.", + "output": "Did your brother go to London last year?" + }, + { + "instruction": "Rewrite the sentence below so that it is grammatically correct.", + "input": "Him and I seen the movie.", + "output": "He and I saw the movie." + }, + { + "instruction": "Edit the given text to ensure all plural nouns are spelled correctly.", + "input": "The boxes was stacked neatly.", + "output": "The boxes were stacked neatly." + }, + { + "instruction": "Rewrite this English sentence into passive voice.", + "input": "Alex fixed the car.", + "output": "The car was fixed by Alex." + }, + { + "instruction": "Rewrite this sentence in the form of a question.", + "input": "This is the biggest tree in the park.", + "output": "Is this the biggest tree in the park?" + }, + { + "instruction": "Edit the text so that it is grammatically correct.", + "input": "He don't like going there", + "output": "He doesn't like going there." + }, + { + "instruction": "Identify the hypernym for the word \"car\".", + "input": "", + "output": "The hypernym for the word \"car\" is \"vehicle\"." + }, + { + "instruction": "Construct a sentence using the words \"achieve\", \"goal\", and \"perseverance\".", + "input": "", + "output": "With perseverance, you can achieve any goal." + }, + { + "instruction": "Rewrite this statement as an imperative sentence.", + "input": "You should check the document for errors.", + "output": "Check the document for errors." + }, + { + "instruction": "Capitalize each word in the sentence.", + "input": "i love watching the sunrise", + "output": "I Love Watching The Sunrise" + }, + { + "instruction": "Create a new sentence by combining the two sentences.", + "input": "He likes to swim. He goes to the pool every day.", + "output": "He likes to swim and goes to the pool every day." + }, + { + "instruction": "Combine the two ideas in the sentence.", + "input": "Learning a new language is useful. It can be challenging.", + "output": "Learning a new language is useful but can be challenging." + }, + { + "instruction": "Reword the sentence to use other words without changing the meaning.", + "input": "It rained heavily during the night.", + "output": "The night experienced significant rainfall." + }, + { + "instruction": "Rewrite the sentence to use a negative adverb.", + "input": "She always remembers to call.", + "output": "She never forgets to call." + }, + { + "instruction": "Categorize the given list of animals.", + "input": "Shark, Dolphin, Trout", + "output": "Fish: Shark, Trout\nMammals: Dolphin" + }, + { + "instruction": "Rewrite the sentence using a synonym.", + "input": "It's cold outside.", + "output": "It's chilly outside." + }, + { + "instruction": "Change the given sentence into passive voice.", + "input": "They celebrate the festival annually.", + "output": "The festival is celebrated annually." + }, + { + "instruction": "Generate a question with the keyword “bacteria”", + "input": "", + "output": "How do bacteria affect human health?" + }, + { + "instruction": "Take the following noun and turn it into a verb.", + "input": "Light", + "output": "To light." + }, + { + "instruction": "Evaluate the following phrase by transforming it into the spelling given.", + "input": "freind --> friend", + "output": "The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\"." + }, + { + "instruction": "Select the noun phrase in the following sentence.", + "input": "He enjoyed a delicious cake at the party.", + "output": "The noun phrases in the given sentence are \"a delicious cake\" and \"the party.\"" + }, + { + "instruction": "Rewrite the following phrase in the present tense.", + "input": "They had played soccer all afternoon.", + "output": "They play soccer all afternoon." + }, + { + "instruction": "Rewrite the following phrase as a question.", + "input": "We must finish the project by tomorrow.", + "output": "Must we finish the project by tomorrow?" + }, + { + "instruction": "Transform the sentence into a negative statement.", + "input": "She always arrives on time.", + "output": "She never arrives late." + }, + { + "instruction": "Suggest a simpler word to replace the word in braces.", + "input": "He moved {expeditiously} through the crowd.", + "output": "He moved quickly through the crowd." + }, + { + "instruction": "Generate a five-word phrase using the words \"moon,\" \"crisp,\" and \"night.\"", + "input": "", + "output": "\"Crisp moon illuminates the night.\"" + }, + { + "instruction": "Describe the color blue in 3 words.", + "input": "", + "output": "Deep, calming, cool." + }, + { + "instruction": "Classify the following species as dogs or cats.", + "input": "Siamese", + "output": "Siamese is a cat." + }, + { + "instruction": "Classify an input string as either a noun or a verb.", + "input": "jump", + "output": "\"jump\" can be classified as a verb." + }, + { + "instruction": "Insert the correct punctuation at the provided places in the following sentence.", + "input": "Where are you going", + "output": "Where are you going?" + }, + { + "instruction": "Convert 10 feet to meters.", + "input": "", + "output": "10 feet is equal to approximately 3.048 meters." + }, + { + "instruction": "Simplify the given sentence.", + "input": "The teacher explained that the reason we were studying this was because it is important.", + "output": "The teacher explained that we were studying this because it's important." + }, + { + "instruction": "Determine whether the given two words are synonyms or antonyms.", + "input": "Open - Close", + "output": "Open and Close are antonyms." + }, + { + "instruction": "Change this sentence grammatically.", + "input": "They has seen the movie.", + "output": "They have seen the movie." + }, + { + "instruction": "Generate a past-tense verb that describes a person singing.", + "input": "", + "output": "Sang." + }, + { + "instruction": "Find the synonym for the given verb.", + "input": "Begin", + "output": "Commence" + }, + { + "instruction": "Name 3 different animals that are active during the day.", + "input": "", + "output": "1. Squirrel\n2. Eagle\n3. Tiger" + }, + { + "instruction": "Fill in the blanks with appropriate words.", + "input": "The sky was filled with colorful ________ at sunset.", + "output": "The sky was filled with colorful clouds at sunset." + }, + { + "instruction": "Categorize the following items into one of the following categories: appliance, clothing, or food.", + "input": "Microwave, T-shirt, Bread", + "output": "Appliance: Microwave\nClothing: T-shirt\nFood: Bread" + }, + { + "instruction": "Edit the sentence: \"We enjoys watching movies\"", + "input": "", + "output": "We enjoy watching movies." + }, + { + "instruction": "Edit the given sentence to make it grammatically correct.", + "input": "She dont have any money", + "output": "She doesn't have any money." + }, + { + "instruction": "Classify the following animal as carnivore, herbivore, or omnivore.", + "input": "Deer", + "output": "Herbivore" + }, + { + "instruction": "Fix the following grammar mistake.", + "input": "He want to go to the movies, but he has no money.", + "output": "He wants to go to the movies, but he has no money." + }, + { + "instruction": "Automatically correct the given sentence.", + "input": "I want to go to london.", + "output": "I want to go to London." + }, + { + "instruction": "Generate a random sequence of 10 digits.", + "input": "", + "output": "Here is a randomly generated sequence of 10 digits: 9274036185." + }, + { + "instruction": "Generate an antonym for the word 'optimistic'.", + "input": "", + "output": "An antonym for the word 'optimistic' could be 'pessimistic'." + }, + { + "instruction": "Edit the text for grammar and punctuation.", + "input": "today was amazing I went hiking and saw a bear", + "output": "\"Today was amazing. I went hiking and saw a bear.\"" + }, + { + "instruction": "Rewrite this sentence to eliminate the passive voice.", + "input": "The cake was made by Susan.", + "output": "Susan made the cake." + }, + { + "instruction": "Sort the following array in ascending order.", + "input": "[9, 2, 7, 4]", + "output": "Here is the sorted array in ascending order: [2, 4, 7, 9]." + }, + { + "instruction": "Add the correct verb to the sentence:\nThe students _________ the quiz answers.", + "input": "", + "output": "The students discussed the quiz answers." + }, + { + "instruction": "Put the following words in alphabetical order: kitten, zebra, alligator.", + "input": "", + "output": "Alligator, kitten, zebra." + }, + { + "instruction": "Delete the redundant words in the following sentence:\nWe went to see to the mountains.", + "input": "", + "output": "We went to see the mountains." + }, + { + "instruction": "Rewrite this sentence using appropriate grammar rules:", + "input": "he saw a old man", + "output": "He saw an old man." + }, + { + "instruction": "Find the next 3 terms in the following sequence: 3, 5, 7, 9", + "input": "", + "output": "The next three terms in the sequence are: 11, 13, 15." + }, + { + "instruction": "Reverse the order of the given phrase.", + "input": "sun and moon", + "output": "moon and sun" + } +] diff --git a/Codes/ch07/02_dataset-utilities/requirements-extra.txt b/Codes/ch07/02_dataset-utilities/requirements-extra.txt new file mode 100644 index 0000000..f4f7970 --- /dev/null +++ b/Codes/ch07/02_dataset-utilities/requirements-extra.txt @@ -0,0 +1,3 @@ +openai>=1.30.3 +scikit-learn>=1.3.1 +tqdm>=4.65.0 \ No newline at end of file diff --git a/Codes/ch07/03_model-evaluation/README.md b/Codes/ch07/03_model-evaluation/README.md new file mode 100644 index 0000000..ea1a2ce --- /dev/null +++ b/Codes/ch07/03_model-evaluation/README.md @@ -0,0 +1,26 @@ +# Chapter 7: Finetuning to Follow Instructions + +This folder contains utility code that can be used for model evaluation. + + + +  +## Evaluating Instruction Responses Using the OpenAI API + + +- The [llm-instruction-eval-openai.ipynb](llm-instruction-eval-openai.ipynb) notebook uses OpenAI's GPT-4 to evaluate responses generated by instruction finetuned models. It works with a JSON file in the following format: + +```python +{ + "instruction": "What is the atomic number of helium?", + "input": "", + "output": "The atomic number of helium is 2.", # <-- The target given in the test set + "model 1 response": "\nThe atomic number of helium is 2.0.", # <-- Response by an LLM + "model 2 response": "\nThe atomic number of helium is 3." # <-- Response by a 2nd LLM +}, +``` + +  +## Evaluating Instruction Responses Locally Using Ollama + +- The [llm-instruction-eval-ollama.ipynb](llm-instruction-eval-ollama.ipynb) notebook offers an alternative to the one above, utilizing a locally downloaded Llama 3 model via Ollama. \ No newline at end of file diff --git a/Codes/ch07/03_model-evaluation/config.json b/Codes/ch07/03_model-evaluation/config.json new file mode 100644 index 0000000..a01cb47 --- /dev/null +++ b/Codes/ch07/03_model-evaluation/config.json @@ -0,0 +1,4 @@ +{ + "OPENAI_API_KEY": "sk-...", + "_comment": "Enter your API key from https://platform.openai.com/api-keys" +} diff --git a/Codes/ch07/03_model-evaluation/eval-example-data.json b/Codes/ch07/03_model-evaluation/eval-example-data.json new file mode 100644 index 0000000..cc7fe10 --- /dev/null +++ b/Codes/ch07/03_model-evaluation/eval-example-data.json @@ -0,0 +1,702 @@ +[ + { + "instruction": "Calculate the hypotenuse of a right triangle with legs of 6 cm and 8 cm.", + "input": "", + "output": "The hypotenuse of the triangle is 10 cm.", + "model 1 response": "\nThe hypotenuse of the triangle is 3 cm.", + "model 2 response": "\nThe hypotenuse of the triangle is 12 cm." + }, + { + "instruction": "Name 3 different animals that are active during the day.", + "input": "", + "output": "1. Squirrel\n2. Eagle\n3. Tiger", + "model 1 response": "\n1. Squirrel\n2. Tiger\n3. Eagle\n4. Cobra\n5. Tiger\n6. Cobra", + "model 2 response": "\n1. Squirrel\n2. Eagle\n3. Tiger" + }, + { + "instruction": "Rewrite the following sentence in a more formal way:\n\"I need to find out what's wrong.\"", + "input": "", + "output": "I must ascertain what is incorrect.", + "model 1 response": "\nWhat is incorrect?", + "model 2 response": "\nI need to find out what's wrong." + }, + { + "instruction": "Identify the interjection in the sentence.", + "input": "Wow, that was an amazing trick!", + "output": "The interjection in the sentence is 'Wow'.", + "model 1 response": "\nThe interjection in the sentence is 'Wow'.", + "model 2 response": "\nThe interjection in the sentence is 'Wow'." + }, + { + "instruction": "Identify the type of sentence.", + "input": "Did you finish the report?", + "output": "The type of sentence is interrogative.", + "model 1 response": "\nThe type of sentence is exclamatory.", + "model 2 response": "\nThe type of sentence is exclamatory." + }, + { + "instruction": "Convert the following sentence into future tense.", + "input": "She walks to school.", + "output": "She will walk to school.", + "model 1 response": "\nShe will walk to school.", + "model 2 response": "\nShe will walk to school." + }, + { + "instruction": "Convert 500 grams to ounces.", + "input": "", + "output": "500 grams is approximately 17.64 ounces.", + "model 1 response": "\n500 grams is approximately 99.21 ounces.", + "model 2 response": "\n500 grams is approximately 0.05 ounces." + }, + { + "instruction": "Edit the following sentence to remove all passive voice.", + "input": "The song was composed by the artist.", + "output": "The artist composed the song.", + "model 1 response": "\nThe artist composed the song.", + "model 2 response": "\nThe artist composed the song." + }, + { + "instruction": "Provide a word that rhymes with \"care.\"", + "input": "", + "output": "A word that rhymes with \"care\" is \"fare.\"", + "model 1 response": "\nA word that rhymes with \"care\" is \"bare.\"", + "model 2 response": "\nA word that rhymes with \"care\" is \"care-free.\"" + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The meal was not delicious.", + "output": "The meal could use some improvement.", + "model 1 response": "\nThe meal could use some improvement.", + "model 2 response": "\nThe meal was delicious." + }, + { + "instruction": "Generate a sentence using the word 'elusive'.", + "input": "", + "output": "The meaning of the poem was elusive, escaping easy interpretation.", + "model 1 response": "\nThe poem was elusive, elusive, and elusive.", + "model 2 response": "\nThe elusive object that is so elusive that it is impossible to describe is the elusive cat." + }, + { + "instruction": "Arrange the given numbers in descending order.", + "input": "7, 10, 3, 6, 9", + "output": "10, 9, 7, 6, 3.", + "model 1 response": "\n10 numbers in descending order: 3, 6, 3.", + "model 2 response": "\nThe numbers in descending order are: 3, 6, 9." + }, + { + "instruction": "Convert this direct speech to indirect speech.", + "input": "'I am tired,' he said.", + "output": "He said that he was tired.", + "model 1 response": "\nHe said that he was tired.", + "model 2 response": "\nHe said that he was tired." + }, + { + "instruction": "What is the capital of Turkey?", + "input": "", + "output": "The capital of Turkey is Ankara.", + "model 1 response": "\nThe capital of Turkey is Ankara.", + "model 2 response": "\nThe capital of Turkey is Ankara." + }, + { + "instruction": "Convert the sentence to future tense.", + "input": "They play football every Sunday.", + "output": "They will play football every Sunday.", + "model 1 response": "\nThey will play football every Sunday.", + "model 2 response": "\nThey will play football every Sunday." + }, + { + "instruction": "Provide a synonym for 'quick'.", + "input": "", + "output": "A synonym for 'quick' is 'fast'.", + "model 1 response": "\nA synonym for 'quick' is 'fast'.", + "model 2 response": "\nA synonym for 'quick' is 'fast'." + }, + { + "instruction": "What is a synonym of 'luminous'?", + "input": "", + "output": "A synonym for 'luminous' is 'bright'.", + "model 1 response": "\nA synonym for 'luminous' is 'bright'.", + "model 2 response": "\nA synonym for 'luminous' is 'bright'." + }, + { + "instruction": "Name the process by which plants lose water vapor through their leaves.", + "input": "", + "output": "The process by which plants lose water vapor through their leaves is called transpiration.", + "model 1 response": "\nThe process is called evaporation.", + "model 2 response": "\nThe process by which plants lose water vapor through their leaves is called transpiration." + }, + { + "instruction": "Identify the verb in the sentence.", + "input": "Birds fly south for the winter.", + "output": "The verb in the sentence is 'fly'.", + "model 1 response": "\nThe verb in the sentence is \"fly.\"", + "model 2 response": "\nThe verb in the sentence is 'fly'." + }, + { + "instruction": "What type of bond is formed when one atom donates an electron to another atom?", + "input": "", + "output": "The type of bond formed when one atom donates an electron to another atom is an ionic bond.", + "model 1 response": "\nThe type of bond formed when one atom donates an electron to another atom is an ionic bond.", + "model 2 response": "\nThe type of bond formed when one atom donates an electron to another atom is ionic." + }, + { + "instruction": "Rewrite the following sentence to include a simile.", + "input": "The car moved fast.", + "output": "The car moved as fast as a cheetah.", + "model 1 response": "\nThe car moved as fast as a cheetah.", + "model 2 response": "\nThe car moved at a fast pace." + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Necesary", + "output": "The correct spelling is \"Necessary\".", + "model 1 response": "\nThe correct spelling is \"Necesary\".", + "model 2 response": "\nThe correct spelling of the word \"Necessary\" is \"Necessary\"." + }, + { + "instruction": "Rewrite this sentence in the active voice.", + "input": "A novel was written by Mary.", + "output": "Mary wrote a novel.", + "model 1 response": "\nMary wrote a novel.", + "model 2 response": "\nMary wrote a novel." + }, + { + "instruction": "rewrite this sentence as a question.", + "input": "Your brother went to London last year.", + "output": "Did your brother go to London last year?", + "model 1 response": "\nDid your brother go to London last year?", + "model 2 response": "\nDid your brother go to London last year?" + }, + { + "instruction": "Explain what a haiku is.", + "input": "", + "output": "A haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5.", + "model 1 response": "\nA haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5.", + "model 2 response": "\nA haiku is a form of traditional Japanese poetry that consists of three lines that are composed of three syllables." + }, + { + "instruction": "Split the following sentence into two declarative sentences: \"The book is interesting but the cover is dull\".", + "input": "", + "output": "The book is interesting.\nThe cover is dull.", + "model 1 response": "\nThe book is interesting.\nThe cover is dull.", + "model 2 response": "\nThe book is dull." + }, + { + "instruction": "What is the antonym of 'sparse'?", + "input": "", + "output": "The antonym of 'sparse' is 'dense'.", + "model 1 response": "\nThe antonym of 'sparse' is 'dense'.", + "model 2 response": "\nThe antonym of 'sparse' is 'dense'." + }, + { + "instruction": "Identify the type of sentence.", + "input": "What a beautiful view!", + "output": "The type of sentence is exclamatory.", + "model 1 response": "\nThe type of sentence is exclamatory.", + "model 2 response": "\nThe type of sentence is subjunctive." + }, + { + "instruction": "What is the boiling point of mercury in Celsius?", + "input": "", + "output": "The boiling point of mercury is 356.7 degrees Celsius.", + "model 1 response": "\nThe boiling point of mercury is approximately 100 degrees Celsius.", + "model 2 response": "\nThe boiling point of mercury is 100 degrees Celsius." + }, + { + "instruction": "Define the term 'kinetic energy'.", + "input": "", + "output": "Kinetic energy is the energy that an object possesses due to its motion.", + "model 1 response": "\nKinetic energy is the energy that an object possesses due to its motion.", + "model 2 response": "\nKinetic energy is the energy that an object possesses due to its motion." + }, + { + "instruction": "Change the following sentence to past perfect tense.", + "input": "They finish the game.", + "output": "They had finished the game.", + "model 1 response": "\nThey had finished the game.", + "model 2 response": "\nThey will finish the game." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Mercury, oxygen, wood", + "output": "Mercury - Liquid\nOxygen - Gas\nWood - Solid", + "model 1 response": "\nMercury: Solid\nOxygen: Gas\nWood: Solid\nPlants: Vegetable\nMinerals: Iron\nMinerals: Gold\nMinerals: Coal\nMinerals: Bacteria: E. coli\nFruits: Cherries\nMinerals: Wheat\nMinerals: Vinegar\nMinerals: Fish\nMinerals: Mercury\nMinerals: Oxygen\nMinerals: Carbon\nMinerals: Wood\nMinerals: Gas\nMinerals: Carbon\nMinerals: Wood\nMinerals: Solid\nMinerals: Gas\nMinerals: Solid\nMinerals: Gas\nMinerals: Solid\nMinerals: Solid\nMinerals: Gas\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\nMinerals: Solid\n", + "model 2 response": "\nMercury is a solid. Oxygen is a liquid. Wood is a gas." + }, + { + "instruction": "What is the synonym for 'enormous'?", + "input": "", + "output": "A synonym for 'enormous' is 'gigantic'.", + "model 1 response": "\nThe synonym for 'enormous' is 'gigantic'.", + "model 2 response": "\nThe synonym for 'enormous' is 'gigantic'." + }, + { + "instruction": "Translate the term 'sustainable development' into German.", + "input": "", + "output": "The German translation of 'sustainable development' is 'nachhaltige Entwicklung'.", + "model 1 response": "\nThe German translation of 'sustainable development' is 'nachhaltige Entwicklung'.", + "model 2 response": "\nThe German translation of 'sustainable development' is 'sustainable development f\u00fcr das Entwicklung'." + }, + { + "instruction": "Name three essential elements found in nature.", + "input": "", + "output": "1. Carbon\n2. Oxygen\n3. Iron", + "model 1 response": "\n1. Carbon\n2. Oxygen\n3. Oxygen\n### Response:\n1. Carbon\n2. Oxygen\n3. Oxygen", + "model 2 response": "\n1. Carbon\n2. Iron\n3. Titanium" + }, + { + "instruction": "Rewrite the following phrase as a question.", + "input": "We must finish the project by tomorrow.", + "output": "Must we finish the project by tomorrow?", + "model 1 response": "\nMust we finish the project by tomorrow?", + "model 2 response": "\nWe must finish the project by tomorrow." + }, + { + "instruction": "Combine the given sentences into one cohesive sentence.", + "input": "It is raining. I cannot go to the park.", + "output": "Since it is raining, I cannot go to the park.", + "model 1 response": "\nIt is raining. I cannot go to the park.", + "model 2 response": "\nIt is raining. I cannot go to the park." + }, + { + "instruction": "Calculate the volume of a cube with side length of 3 meters.", + "input": "", + "output": "The volume of the cube is 27 cubic meters.", + "model 1 response": "\nThe volume of the cube is approximately 3 cubic meters.", + "model 2 response": "\nThe volume of the cube is approximately 3.5 cubic meters." + }, + { + "instruction": "Re-word this sentence using an active voice.", + "input": "The decision was made by the team.", + "output": "The team made the decision.", + "model 1 response": "\nThe team made the decision.", + "model 2 response": "\nThe team made the decision." + }, + { + "instruction": "Convert the sentence to use an indefinite pronoun.", + "input": "Someone left a message.", + "output": "A message was left by someone.", + "model 1 response": "\nA message was left by someone.", + "model 2 response": "\nA message was left by someone." + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer crashed.", + "output": "The statement \"My computer crashed\" can be classified as negative.", + "model 1 response": "\nThe statement \"My computer crashed\" can be classified as negative.", + "model 2 response": "\nThe statement \"My computer crashed\" is classified as negative." + }, + { + "instruction": "What is the derivative of x^2?", + "input": "", + "output": "The derivative of x^2 is 2x.", + "model 1 response": "\nThe derivative of x^2 is 2x.", + "model 2 response": "\nThe derivative of x^2 is 2x." + }, + { + "instruction": "Fill in the blanks with appropriate words.", + "input": "The sky was filled with colorful ________ at sunset.", + "output": "The sky was filled with colorful clouds at sunset.", + "model 1 response": "\nThe sky was filled with colorful clouds at sunset.", + "model 2 response": "\nThe sky was filled with colorful clouds at sunset." + }, + { + "instruction": "Change the sentence from passive to active voice.", + "input": "The song was sung by the choir.", + "output": "The choir sang the song.", + "model 1 response": "\nThe choir sang the song.", + "model 2 response": "\nThe choir sang the song." + }, + { + "instruction": "What is the atomic mass of carbon?", + "input": "", + "output": "The atomic mass of carbon is approximately 12.01 atomic mass units.", + "model 1 response": "\nThe atomic mass of carbon is approximately 3.01 atomic mass units.", + "model 2 response": "\nThe atomic mass of carbon is approximately 3.0 atomic mass units." + }, + { + "instruction": "Convert the sentence to present perfect tense: 'He wins the race.'", + "input": "He wins the race.", + "output": "He has won the race.", + "model 1 response": "\nHe has won the race.", + "model 2 response": "\nHe will win the race." + }, + { + "instruction": "Rewrite the following sentence to include an idiom.", + "input": "It started raining suddenly.", + "output": "It started raining cats and dogs.", + "model 1 response": "\nIt started raining cats and dogs.", + "model 2 response": "\nIt started raining suddenly." + }, + { + "instruction": "What is the atomic number of helium?", + "input": "", + "output": "The atomic number of helium is 2.", + "model 1 response": "\nThe atomic number of helium is 2.0.", + "model 2 response": "\nThe atomic number of helium is 3." + }, + { + "instruction": "Name three types of clouds.", + "input": "", + "output": "Three types of clouds are cumulus, stratus, and cirrus.", + "model 1 response": "\nThree types of clouds are cirrus, cirrus-like, and cirrus-like-ring.", + "model 2 response": "\nThree types of clouds are cumulus, stratus, and cirrus." + }, + { + "instruction": "Identify the onomatopoeia in the sentence.", + "input": "The bees buzzed as they flew past.", + "output": "The onomatopoeia in the sentence is 'buzzed'.", + "model 1 response": "\nThe onomatopoeia in the sentence is that the bees buzzed as they flew past.", + "model 2 response": "\nThe onomatopoeia in the sentence is 'buzzed'." + }, + { + "instruction": "Name the process by which water changes from liquid to gas.", + "input": "", + "output": "The process by which water changes from liquid to gas is called evaporation.", + "model 1 response": "\nThe process is called evaporation.", + "model 2 response": "\nThe process by which water changes from liquid to gas is called evaporation." + }, + { + "instruction": "Find the synonym for the given verb.", + "input": "Begin", + "output": "Commence", + "model 1 response": "\nCommence", + "model 2 response": "\nBegin\n\n### Input:\nbegin\n\n### Response:\nBegin\n\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin\n### Response:\nbegin\n### Input:\nbegin" + }, + { + "instruction": "Classify the following numbers as prime or composite.", + "input": ": 11, 14, 19.", + "output": "Prime numbers: 11, 19\nComposite numbers: 14", + "model 1 response": "\nPrime numbers: 11, 19\n\nComposite numbers: 14\n\n### Response:\nPrime numbers: 11, 19\nComposite numbers: 14", + "model 2 response": "\nPrime numbers: 11, 19, 19. Composite numbers: 14, 19, 19." + }, + { + "instruction": "Classify the item as either a grain or a fruit.", + "input": "Wheat", + "output": "Grain", + "model 1 response": "\nGrain\n\n### Response:\nGrain\n### Response:\nWheat\n### Response:\nGrain", + "model 2 response": "\nGrain" + }, + { + "instruction": "Provide the plural form of 'thesis'.", + "input": "", + "output": "The plural form of 'thesis' is 'theses'.", + "model 1 response": "\nThe plural form of 'thesis' is 'theses'.", + "model 2 response": "\nThe plural form of 'thesis' is 'thesis'." + }, + { + "instruction": "Generate three verbs that mean the same as \"to run.\"", + "input": "", + "output": "1. To sprint\n2. To dash\n3. To jog", + "model 1 response": "\n1. To sprint\n2. To dash\n3. To dash\n4. To dash\n5. To dash\n6. To dash", + "model 2 response": "\n1. To sprint\n2. To dash\n3. To jog" + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just returned from a vacation in Spain.", + "output": "How was your vacation in Spain?", + "model 1 response": "\nHow was your vacation in Spain?", + "model 2 response": "\nWhat was your vacation in Spain?" + }, + { + "instruction": "What is the normal boiling point of water?", + "input": "", + "output": "The normal boiling point of water is 100 degrees Celsius.", + "model 1 response": "\nThe normal boiling point of water is 100 degrees Celsius.", + "model 2 response": "\nThe normal boiling point of water is 100 degrees Celsius." + }, + { + "instruction": "Convert the sentence to active voice.", + "input": "The book was read by many.", + "output": "Many people read the book.", + "model 1 response": "\nMany people read the book.", + "model 2 response": "\nMany people read the book." + }, + { + "instruction": "Identify the tense used in the sentence.", + "input": "They are watching a movie.", + "output": "The tense used in the sentence \"They are watching a movie\" is the present continuous tense.", + "model 1 response": "\nThe tense used in the sentence is future continuous.", + "model 2 response": "\nThe tense used in the sentence is future continuous." + }, + { + "instruction": "Explain what GDP stands for and what it measures.", + "input": "", + "output": "GDP stands for Gross Domestic Product, which measures the total value of all goods and services produced over a specific time period within a country's borders.", + "model 1 response": "\nGDP stands for Gross Domestic Product, which measures the total value of all goods and services produced over a specific time period within a country's borders.", + "model 2 response": "\nGDP stands for Gross Domestic Product, which measures the total value of all goods and services produced over a specific time period within a country's borders." + }, + { + "instruction": "Identify the verb in the following sentence", + "input": "The sun shines brightly.", + "output": "The verb in the sentence is 'shines'.", + "model 1 response": "\nThe verb in the sentence is \"shines.\"", + "model 2 response": "\nThe verb in the sentence is 'shines'." + }, + { + "instruction": "Rewrite the following sentence to include a metaphor.'", + "input": "Life can be hard.", + "output": "Life is a rocky road.", + "model 1 response": "\nLife is a rocky road.", + "model 2 response": "\nLife is a rocky road." + }, + { + "instruction": "What is the derivative of the function f(x) = x^4?", + "input": "", + "output": "The derivative of the function f(x) = x^4 is f'(x) = 4x^3.", + "model 1 response": "\nThe derivative of the function f(x) = x^4 is f'(x) = 4x^3.", + "model 2 response": "\nThe derivative of the function f(x) = x^4 is f'(x) = 4x." + }, + { + "instruction": "Convert the temperature from Celsius to Kelvin.", + "input": "25 degrees Celsius", + "output": "25 degrees Celsius is 298.15 Kelvin.", + "model 1 response": "\n25 degrees Celsius is approximately 298.15 Kelvin.", + "model 2 response": "\n25 degrees Celsius is approximately 298.15 Kelvin." + }, + { + "instruction": "Rewrite the following sentence to include a hyperbole: 'I'm very hungry.'", + "input": "I'm very hungry.", + "output": "I'm so hungry I could eat a horse.", + "model 1 response": "\nI'm very hungry.", + "model 2 response": "\nI'm very hungry." + }, + { + "instruction": "List the ingredients required to make a basic salad.", + "input": "", + "output": "Ingredients for a basic salad typically include lettuce, tomatoes, cucumbers, and dressing.", + "model 1 response": "\nIngredients for a basic salad typically include lettuce, tomatoes, cucumbers, and cucumbers.", + "model 2 response": "\nIngredients for a basic salad typically include lettuce, tomatoes, dressing, and dressingings." + }, + { + "instruction": "Select the noun phrase in the following sentence.", + "input": "He enjoyed a delicious cake at the party.", + "output": "The noun phrases in the given sentence are \"a delicious cake\" and \"the party.\"", + "model 1 response": "\nThe noun phrases in the sentence are \"a delicious cake\" and \"the party.\"", + "model 2 response": "\nThe noun phrases in the sentence are \"a delicious cake\" and \"the party.\"" + }, + { + "instruction": "What is the contraction for \"will not\"?", + "input": "", + "output": "The contraction for \"will not\" is \"won't\".", + "model 1 response": "\nThe contraction for \"will not\" is \"won't\".", + "model 2 response": "\nThe contraction for \"will not\" is \"won't\"." + }, + { + "instruction": "Generate a sentence using the word 'elucidate'.", + "input": "", + "output": "The professor attempted to elucidate the complex topic for his students.", + "model 1 response": "\nThe professor elucidated the complex topic for his students.", + "model 2 response": "\nThe professor elucidated the topic of the class." + }, + { + "instruction": "Convert this sentence to passive voice", + "input": "The chef cooked a delicious meal.", + "output": "A delicious meal was cooked by the chef.", + "model 1 response": "\nA delicious meal was cooked by the chef.", + "model 2 response": "\nThe chef cooked a delicious meal." + }, + { + "instruction": "What is the pH of lemon juice?", + "input": "", + "output": "The pH of lemon juice is approximately 2.0.", + "model 1 response": "\nThe pH of lemon juice is approximately 2.0.", + "model 2 response": "\nThe pH of lemon juice is typically around 3.0." + }, + { + "instruction": "What is the capital of Norway?", + "input": "", + "output": "The capital of Norway is Oslo.", + "model 1 response": "\nThe capital of Norway is Oslo.", + "model 2 response": "\nThe capital of Norway is Oslo." + }, + { + "instruction": "Generate a question with the keyword \u201cbacteria\u201d", + "input": "", + "output": "How do bacteria affect human health?", + "model 1 response": "\nWhat is the pH of baking soda solution?\n\n### Response:\nThe pH of a baking soda solution is typically around 2.0.", + "model 2 response": "\nWhat is the pH of a neutral solution at 25\u00b0C?\n\n### Input:\nThe pH of a neutral solution at 25\u00b0C is typically around 3.0." + }, + { + "instruction": "Generate a sentence using the word 'spontaneous'.", + "input": "", + "output": "His decision to travel was completely spontaneous, made on a whim.", + "model 1 response": "\nHis decision to travel was completely spontaneous, made on a whim.", + "model 2 response": "\nThe decision was made by the team." + }, + { + "instruction": "Convert the sentence to future perfect tense.", + "input": "They close the store.", + "output": "They will have closed the store.", + "model 1 response": "\nThey will have closed the store.", + "model 2 response": "\nThey will have closed the store." + }, + { + "instruction": "Provide the past participle form of 'break'.", + "input": "", + "output": "The past participle form of 'break' is 'broken'.", + "model 1 response": "\nThe past participle form of 'break' is 'broken'.", + "model 2 response": "\nThe past participle form of 'break' is 'broken'." + }, + { + "instruction": "Provide the correct contraction for 'they are'.", + "input": "", + "output": "The contraction for 'they are' is 'they're'.", + "model 1 response": "\nThe contraction for 'they are' is 'they're'.", + "model 2 response": "\nThe contraction for 'they are' is 'they're'." + }, + { + "instruction": "Convert the temperature from Fahrenheit to Celsius: 68\u00b0F.", + "input": "68 degrees Fahrenheit", + "output": "68 degrees Fahrenheit is approximately 20 degrees Celsius.", + "model 1 response": "\n68 degrees Fahrenheit is approximately 37 degrees Celsius.", + "model 2 response": "\n68 degrees Fahrenheit is approximately 37 degrees Celsius." + }, + { + "instruction": "Turn the following statement into a question.", + "input": "The Golden State Warriors are a professional basketball team based in San Francisco.", + "output": "Are the Golden State Warriors a professional basketball team based in San Francisco?", + "model 1 response": "\nAre the Golden State Warriors a professional basketball team based in San Francisco?", + "model 2 response": "\nAre the Golden State Warriors a professional basketball team based in San Francisco?" + }, + { + "instruction": "What is the derivative of the function f(x) = 3x^3?", + "input": "", + "output": "The derivative of the function f(x) = 3x^3 is f'(x) = 9x^2.", + "model 1 response": "\nThe derivative of the function f(x) = 3x^3 is f'(x) = 6x.", + "model 2 response": "\nThe derivative of the function f(x) = 3x^3 is f'(x) = 4x." + }, + { + "instruction": "Rewrite the following phrase in the present tense.", + "input": "They had played soccer all afternoon.", + "output": "They play soccer all afternoon.", + "model 1 response": "\nThey play soccer all afternoon.", + "model 2 response": "\nThey played soccer all afternoon." + }, + { + "instruction": "Rewrite this sentence to start with \"Despite\": She managed to finish her work, although she was tired.", + "input": "", + "output": "Despite being tired, she managed to finish her work.", + "model 1 response": "\nDespite being tired, she managed to finish her work.", + "model 2 response": "\nDespite being tired, she managed to finish her work." + }, + { + "instruction": "Convert the sentence to future continuous tense.", + "input": "She plays the piano.", + "output": "She will be playing the piano.", + "model 1 response": "\nShe will play the piano.", + "model 2 response": "\nShe will be playing the piano." + }, + { + "instruction": "Capitalize each word in the sentence.", + "input": "i love watching the sunrise", + "output": "I Love Watching The Sunrise", + "model 1 response": "\nI love watching the sunrise.", + "model 2 response": "\nI love watching the sunrise." + }, + { + "instruction": "What is the antonym of 'complicated'?", + "input": "", + "output": "The antonym of 'complicated' is 'simple'.", + "model 1 response": "\nThe antonym of 'complicated' is 'simple'.", + "model 2 response": "\nThe antonym of 'complicated' is 'simple'." + }, + { + "instruction": "Sort the following numbers in ascending order.", + "input": "5, 18, 3, 11.", + "output": "The numbers in ascending order would be: 3, 5, 11, 18.", + "model 1 response": "\nThe numbers in ascending order are: 3, 5, 11.", + "model 2 response": "\nThe numbers in ascending order are: 3, 11, 5, 18." + }, + { + "instruction": "Identify the theme of the story.", + "input": "The Tortoise and the Hare", + "output": "The theme of the story is that slow and steady wins the race.", + "model 1 response": "\nThe theme of the story is that slow and steady wins the race.", + "model 2 response": "\nThe theme of the story is that the Tortoise and the Hare are the two most popular animals in the world." + }, + { + "instruction": "Correct the verb tense error in the sentence.", + "input": "She go to school every day.", + "output": "The corrected sentence should be: 'She goes to school every day.'", + "model 1 response": "\nThe corrected sentence should be: 'She goes to school every day.'", + "model 2 response": "\nThe corrected sentence should be: 'She goes to school every day.'" + }, + { + "instruction": "Evaluate this sentence for spelling and grammar mistakes.", + "input": "I recieved an invitation to an exciting evenement.", + "output": "There are two errors in the sentence. The corrected sentence should be: \"I received an invitation to an exciting event.\"", + "model 1 response": "\nAn invitation to an exciting event?", + "model 2 response": "\nThe corrected sentence should be: 'I received an invitation to an exciting event.'" + }, + { + "instruction": "Identify the hypernym for the word \"car\".", + "input": "", + "output": "The hypernym for the word \"car\" is \"vehicle\".", + "model 1 response": "\nThe hypernym for the word \"car\" is \"vehicle\".", + "model 2 response": "\nThe hypernym for the word \"car\" is \"vehicle\"." + }, + { + "instruction": "Name an antonym for \"joy.\"", + "input": "", + "output": "An antonym for \"joy\" is \"sorrow.\"", + "model 1 response": "\nAn antonym for \"joy\" is \"sorrow.\"", + "model 2 response": "\nAn antonym for \"joy\" is \"contentment.\"" + }, + { + "instruction": "Convert the sentence into a question", + "input": "She likes ice cream.", + "output": "Does she like ice cream?", + "model 1 response": "\nDoes she like ice cream?", + "model 2 response": "\nDoes she like ice cream?" + }, + { + "instruction": "Rewrite this sentence in the form of a question.", + "input": "This is the biggest tree in the park.", + "output": "Is this the biggest tree in the park?", + "model 1 response": "\nIs this the biggest tree in the park?", + "model 2 response": "\nIs this the biggest tree in the park?" + }, + { + "instruction": "Find a rhyme for the word 'sing'.", + "input": "", + "output": "A word that rhymes with 'sing' is 'ring'.", + "model 1 response": "\nA word that rhymes with 'ring' is 'ring-a'.", + "model 2 response": "\nThe word 'sing' is a derivative of the word 'sing'." + }, + { + "instruction": "Convert this sentence to future perfect tense.", + "input": "She will finish the project.", + "output": "She will have finished the project.", + "model 1 response": "\nShe will have finished the project.", + "model 2 response": "\nShe will have finished the project." + }, + { + "instruction": "What type of number is 0?", + "input": "", + "output": "0 is an integer.", + "model 1 response": "\n0 is an integer.", + "model 2 response": "\n0 is an integer." + }, + { + "instruction": "Classify the following animals", + "input": "Eagle, Shark, Cobra.", + "output": "Birds: Eagle\nFish: Shark\nReptiles: Cobra", + "model 1 response": "\nBirds: Eagle\nFish: Shark\nReptiles: Cobra\nReptiles: Squirrel\nReptiles: Tiger\nReptiles: Cobra\nReptiles: Tiger\nReptiles: Squirrel\nReptiles: Tiger\nReptiles: Cobra\nReptiles: Tiger\nReptiles: Cobra", + "model 2 response": "\nBirds: Eagle\nFish: Shark\nReptiles: Cobra\nPlants: Eagle\nMinerals: Eagle\nFruits: Shark\nFruits: Shark" + }, + { + "instruction": "Take the following noun and turn it into a verb.", + "input": "Light", + "output": "To light.", + "model 1 response": "\nTo light.", + "model 2 response": "\nLight is the most abundant element in the periodic table." + }, + { + "instruction": "Generate a random alphanumeric code of 10 characters.", + "input": "", + "output": "Here is a randomly generated 10-character alphanumeric code: 3F8G6H7JQ2.", + "model 1 response": "\nHere is a randomly generated sequence of 10 characters: 3F8G6H7JQHWQ2JQ3H6H-\n\n### Response:\nHere is a sequence of 10 characters: 3F8G6H7JQHWQ2JQ3H6H-", + "model 2 response": "\nThe alphanumeric code for the word 'happy' is 'happiness'." + } +] \ No newline at end of file diff --git a/Codes/ch07/03_model-evaluation/llm-instruction-eval-ollama.ipynb b/Codes/ch07/03_model-evaluation/llm-instruction-eval-ollama.ipynb new file mode 100644 index 0000000..cc9673f --- /dev/null +++ b/Codes/ch07/03_model-evaluation/llm-instruction-eval-ollama.ipynb @@ -0,0 +1,648 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "136a4efe-fb99-4311-8679-e0a5b6282755", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "b1910a06-e8a3-40ac-8201-ff70615b1ba4", + "metadata": { + "tags": [] + }, + "source": [ + "# Evaluating Instruction Responses Locally Using a Llama 3 Model Via Ollama" + ] + }, + { + "cell_type": "markdown", + "id": "a128651b-f326-4232-a994-42f38b7ed520", + "metadata": {}, + "source": [ + "- This notebook uses an 8-billion-parameter Llama 3 model through ollama to evaluate responses of instruction finetuned LLMs based on a dataset in JSON format that includes the generated model responses, for example:\n", + "\n", + "\n", + "\n", + "```python\n", + "{\n", + " \"instruction\": \"What is the atomic number of helium?\",\n", + " \"input\": \"\",\n", + " \"output\": \"The atomic number of helium is 2.\", # <-- The target given in the test set\n", + " \"model 1 response\": \"\\nThe atomic number of helium is 2.0.\", # <-- Response by an LLM\n", + " \"model 2 response\": \"\\nThe atomic number of helium is 3.\" # <-- Response by a 2nd LLM\n", + "},\n", + "```\n", + "\n", + "- The code doesn't require a GPU and runs on a laptop (it was tested on a M3 MacBook Air)" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "63610acc-db94-437f-8d38-e99dca0299cb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tqdm version: 4.66.4\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\"tqdm\", # Progress bar\n", + " ]\n", + "\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "markdown", + "id": "8bcdcb34-ac75-4f4f-9505-3ce0666c42d5", + "metadata": {}, + "source": [ + "## Installing Ollama and Downloading Llama 3" + ] + }, + { + "cell_type": "markdown", + "id": "5a092280-5462-4709-a3fe-8669a4a8a0a6", + "metadata": {}, + "source": [ + "- Ollama is an application to run LLMs efficiently\n", + "- It is a wrapper around [llama.cpp](https://github.com/ggerganov/llama.cpp), which implements LLMs in pure C/C++ to maximize efficiency\n", + "- Note that it is a tool for using LLMs to generate text (inference), not training or finetuning LLMs\n", + "- Prior to running the code below, install ollama by visiting [https://ollama.com](https://ollama.com) and following the instructions (for instance, clicking on the \"Download\" button and downloading the ollama application for your operating system)" + ] + }, + { + "cell_type": "markdown", + "id": "9558a522-650d-401a-84fc-9fd7b1f39da7", + "metadata": {}, + "source": [ + "- For macOS and Windows users, click on the ollama application you downloaded; if it prompts you to install the command line usage, say \"yes\"\n", + "- Linux users can use the installation command provided on the ollama website\n", + "\n", + "- In general, before we can use ollama from the command line, we have to either start the ollama application or run `ollama serve` in a separate terminal\n", + "\n", + "\n", + "\n", + "\n", + "- With the ollama application or `ollama serve` running, in a different terminal, on the command line, execute the following command to try out the 8-billion-parameter Llama 3 model (the model, which takes up 4.7 GB of storage space, will be automatically downloaded the first time you execute this command)\n", + "\n", + "```bash\n", + "# 8B model\n", + "ollama run llama3\n", + "```\n", + "\n", + "\n", + "The output looks like as follows:\n", + "\n", + "```\n", + "$ ollama run llama3\n", + "pulling manifest \n", + "pulling 6a0746a1ec1a... 100% ▕████████████████▏ 4.7 GB                         \n", + "pulling 4fa551d4f938... 100% ▕████████████████▏  12 KB                         \n", + "pulling 8ab4849b038c... 100% ▕████████████████▏  254 B                         \n", + "pulling 577073ffcc6c... 100% ▕████████████████▏  110 B                         \n", + "pulling 3f8eb4da87fa... 100% ▕████████████████▏  485 B                         \n", + "verifying sha256 digest \n", + "writing manifest \n", + "removing any unused layers \n", + "success \n", + "```\n", + "\n", + "- Note that `llama3` refers to the instruction finetuned 8-billion-parameter Llama 3 model\n", + "\n", + "- Alternatively, you can also use the larger 70-billion-parameter Llama 3 model, if your machine supports it, by replacing `llama3` with `llama3:70b`\n", + "\n", + "- After the download has been completed, you will see a command line prompt that allows you to chat with the model\n", + "\n", + "- Try a prompt like \"What do llamas eat?\", which should return an output similar to the following:\n", + "\n", + "```\n", + ">>> What do llamas eat?\n", + "Llamas are ruminant animals, which means they have a four-chambered \n", + "stomach and eat plants that are high in fiber. In the wild, llamas \n", + "typically feed on:\n", + "1. Grasses: They love to graze on various types of grasses, including tall \n", + "grasses, wheat, oats, and barley.\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "0b5addcb-fc7d-455d-bee9-6cc7a0d684c7", + "metadata": {}, + "source": [ + "- You can end this session using the input `/bye`" + ] + }, + { + "cell_type": "markdown", + "id": "dda155ee-cf36-44d3-b634-20ba8e1ca38a", + "metadata": {}, + "source": [ + "## Using Ollama's REST API" + ] + }, + { + "cell_type": "markdown", + "id": "89343a84-0ddc-42fc-bf50-298a342b93c0", + "metadata": {}, + "source": [ + "- Now, an alternative way to interact with the model is via its REST API in Python via the following function\n", + "- Before you run the next cells in this notebook, make sure that ollama is still running, as described above, via\n", + " - `ollama serve` in a terminal\n", + " - the ollama application\n", + "- Next, run the following code cell to query the model" + ] + }, + { + "cell_type": "markdown", + "id": "16642a48-1cab-40d2-af08-ab8c2fbf5876", + "metadata": {}, + "source": [ + "- First, let's try the API with a simple example to make sure it works as intended:" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "65b0ba76-1fb1-4306-a7c2-8f3bb637ccdb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Llamas are herbivores, which means they primarily feed on plant-based foods. Their diet typically consists of:\n", + "\n", + "1. Grasses: Llamas love to graze on various types of grasses, including tall grasses, short grasses, and even weeds.\n", + "2. Hay: High-quality hay, such as alfalfa or timothy hay, is a staple in a llama's diet. They enjoy the sweet taste and texture of fresh hay.\n", + "3. Grains: Llamas may receive grains like oats, barley, or corn as part of their daily ration. However, it's essential to provide these grains in moderation, as they can be high in calories.\n", + "4. Fruits and vegetables: Llamas enjoy a variety of fruits and veggies, such as apples, carrots, sweet potatoes, and leafy greens like kale or spinach.\n", + "5. Minerals: Llamas require access to mineral supplements, which help maintain their overall health and well-being.\n", + "\n", + "In the wild, llamas might also eat:\n", + "\n", + "1. Leaves: They'll munch on leaves from trees and shrubs, including plants like willow, alder, and birch.\n", + "2. Bark: In some cases, llamas may eat the bark of certain trees, like aspen or cottonwood.\n", + "3. Mosses and lichens: These non-vascular plants can be a tasty snack for llamas.\n", + "\n", + "In captivity, llama owners typically provide a balanced diet that includes a mix of hay, grains, and fruits/vegetables. It's essential to consult with a veterinarian or experienced llama breeder to determine the best feeding plan for your llama.\n" + ] + } + ], + "source": [ + "import urllib.request\n", + "import json\n", + "\n", + "\n", + "def query_model(prompt, model=\"llama3\", url=\"http://localhost:11434/api/chat\"):\n", + " # Create the data payload as a dictionary\n", + " data = {\n", + " \"model\": model,\n", + " \"messages\": [\n", + " {\n", + " \"role\": \"user\",\n", + " \"content\": prompt\n", + " }\n", + " ],\n", + " \"options\": { # Settings below are required for deterministic responses\n", + " \"seed\": 123,\n", + " \"temperature\": 0,\n", + " \"num_ctx\": 2048\n", + " }\n", + " }\n", + "\n", + " # Convert the dictionary to a JSON formatted string and encode it to bytes\n", + " payload = json.dumps(data).encode(\"utf-8\")\n", + "\n", + " # Create a request object, setting the method to POST and adding necessary headers\n", + " request = urllib.request.Request(url, data=payload, method=\"POST\")\n", + " request.add_header(\"Content-Type\", \"application/json\")\n", + "\n", + " # Send the request and capture the response\n", + " response_data = \"\"\n", + " with urllib.request.urlopen(request) as response:\n", + " # Read and decode the response\n", + " while True:\n", + " line = response.readline().decode(\"utf-8\")\n", + " if not line:\n", + " break\n", + " response_json = json.loads(line)\n", + " response_data += response_json[\"message\"][\"content\"]\n", + "\n", + " return response_data\n", + "\n", + "\n", + "result = query_model(\"What do Llamas eat?\")\n", + "print(result)" + ] + }, + { + "cell_type": "markdown", + "id": "162a4739-6f03-4092-a5c2-f57a0b6a4c4d", + "metadata": {}, + "source": [ + "## Load JSON Entries" + ] + }, + { + "cell_type": "markdown", + "id": "ca011a8b-20c5-4101-979e-9b5fccf62f8a", + "metadata": {}, + "source": [ + "- Now, let's get to the data evaluation part\n", + "- Here, we assume that we saved the test dataset and the model responses as a JSON file that we can load as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "8b2d393a-aa92-4190-9d44-44326a6f699b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of entries: 100\n" + ] + } + ], + "source": [ + "json_file = \"eval-example-data.json\"\n", + "\n", + "with open(json_file, \"r\") as file:\n", + " json_data = json.load(file)\n", + "\n", + "print(\"Number of entries:\", len(json_data))" + ] + }, + { + "cell_type": "markdown", + "id": "b6c9751b-59b7-43fe-acc7-14e8daf2fa66", + "metadata": {}, + "source": [ + "- The structure of this file is as follows, where we have the given response in the test dataset (`'output'`) and responses by two different models (`'model 1 response'` and `'model 2 response'`):" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "7222fdc0-5684-4f2b-b741-3e341851359e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'instruction': 'Calculate the hypotenuse of a right triangle with legs of 6 cm and 8 cm.',\n", + " 'input': '',\n", + " 'output': 'The hypotenuse of the triangle is 10 cm.',\n", + " 'model 1 response': '\\nThe hypotenuse of the triangle is 3 cm.',\n", + " 'model 2 response': '\\nThe hypotenuse of the triangle is 12 cm.'}" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "json_data[0]" + ] + }, + { + "cell_type": "markdown", + "id": "fcf0331b-6024-4bba-89a9-a088b14a1046", + "metadata": {}, + "source": [ + "- Below is a small utility function that formats the input for visualization purposes later:" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "43263cd3-e5fb-4ab5-871e-3ad6e7d21a8c", + "metadata": {}, + "outputs": [], + "source": [ + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"Below is an instruction that describes a task. Write a response that \"\n", + " f\"appropriately completes the request.\"\n", + " f\"\\n\\n### Instruction:\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n\\n### Input:\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + " instruction_text + input_text\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "markdown", + "id": "39a55283-7d51-4136-ba60-f799d49f4098", + "metadata": {}, + "source": [ + "- Now, let's try the ollama API to compare the model responses (we only evaluate the first 5 responses for a visual comparison):" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "735cc089-d127-480a-b39d-0782581f0c41", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Dataset response:\n", + ">> The hypotenuse of the triangle is 10 cm.\n", + "\n", + "Model response:\n", + ">> \n", + "The hypotenuse of the triangle is 3 cm.\n", + "\n", + "Score:\n", + ">> I'd score this response as 0 out of 100.\n", + "\n", + "The correct answer is \"The hypotenuse of the triangle is 10 cm.\", not \"3 cm.\". The model failed to accurately calculate the length of the hypotenuse, which is a fundamental concept in geometry and trigonometry.\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> 1. Squirrel\n", + "2. Eagle\n", + "3. Tiger\n", + "\n", + "Model response:\n", + ">> \n", + "1. Squirrel\n", + "2. Tiger\n", + "3. Eagle\n", + "4. Cobra\n", + "5. Tiger\n", + "6. Cobra\n", + "\n", + "Score:\n", + ">> I'd rate this model response as 60 out of 100.\n", + "\n", + "Here's why:\n", + "\n", + "* The model correctly identifies two animals that are active during the day: Squirrel and Eagle.\n", + "* However, it incorrectly includes Tiger twice, which is not a different animal from the original list.\n", + "* Cobra is also an incorrect answer, as it is typically nocturnal or crepuscular (active at twilight).\n", + "* The response does not meet the instruction to provide three different animals that are active during the day.\n", + "\n", + "To achieve a higher score, the model should have provided three unique and correct answers that fit the instruction.\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> I must ascertain what is incorrect.\n", + "\n", + "Model response:\n", + ">> \n", + "What is incorrect?\n", + "\n", + "Score:\n", + ">> A clever test!\n", + "\n", + "Here's my attempt at rewriting the sentence in a more formal way:\n", + "\n", + "\"I require an identification of the issue.\"\n", + "\n", + "Now, let's evaluate the model response \"What is incorrect?\" against the correct output \"I must ascertain what is incorrect.\".\n", + "\n", + "To me, this seems like a completely different question being asked. The original instruction was to rewrite the sentence in a more formal way, and the model response doesn't even attempt to do that. It's asking a new question altogether!\n", + "\n", + "So, I'd score this response a 0 out of 100.\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> The interjection in the sentence is 'Wow'.\n", + "\n", + "Model response:\n", + ">> \n", + "The interjection in the sentence is 'Wow'.\n", + "\n", + "Score:\n", + ">> I'd score this model response as 100.\n", + "\n", + "Here's why:\n", + "\n", + "1. The instruction asks to identify the interjection in the sentence.\n", + "2. The input sentence is provided: \"Wow, that was an amazing trick!\"\n", + "3. The model correctly identifies the interjection as \"Wow\", which is a common English interjection used to express surprise or excitement.\n", + "4. The response accurately answers the question and provides the correct information.\n", + "\n", + "Overall, the model's response perfectly completes the request, making it a 100% accurate answer!\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> The type of sentence is interrogative.\n", + "\n", + "Model response:\n", + ">> \n", + "The type of sentence is exclamatory.\n", + "\n", + "Score:\n", + ">> I'd rate this model response as 20 out of 100.\n", + "\n", + "Here's why:\n", + "\n", + "* The input sentence \"Did you finish the report?\" is indeed an interrogative sentence, which asks a question.\n", + "* The model response says it's exclamatory, which is incorrect. Exclamatory sentences are typically marked by an exclamation mark (!) and express strong emotions or emphasis, whereas this sentence is simply asking a question.\n", + "\n", + "The correct output \"The type of sentence is interrogative.\" is the best possible score (100), while the model response is significantly off the mark, hence the low score.\n", + "\n", + "-------------------------\n" + ] + } + ], + "source": [ + "for entry in json_data[:5]:\n", + " prompt = (f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"score the model response `{entry['model 1 response']}`\"\n", + " f\" on a scale from 0 to 100, where 100 is the best score. \"\n", + " )\n", + " print(\"\\nDataset response:\")\n", + " print(\">>\", entry['output'])\n", + " print(\"\\nModel response:\")\n", + " print(\">>\", entry[\"model 1 response\"])\n", + " print(\"\\nScore:\")\n", + " print(\">>\", query_model(prompt))\n", + " print(\"\\n-------------------------\")" + ] + }, + { + "cell_type": "markdown", + "id": "142dfaa7-429f-4eb0-b74d-ff327f79547a", + "metadata": {}, + "source": [ + "- Note that the responses are very verbose; to quantify which model is better, we only want to return the scores:" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "3552bdfb-7511-42ac-a9ec-da672e2a5468", + "metadata": {}, + "outputs": [], + "source": [ + "from tqdm import tqdm\n", + "\n", + "\n", + "def generate_model_scores(json_data, json_key):\n", + " scores = []\n", + " for entry in tqdm(json_data, desc=\"Scoring entries\"):\n", + " prompt = (\n", + " f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"score the model response `{entry[json_key]}`\"\n", + " f\" on a scale from 0 to 100, where 100 is the best score. \"\n", + " f\"Respond with the integer number only.\"\n", + " )\n", + " score = query_model(prompt)\n", + " try:\n", + " scores.append(int(score))\n", + " except ValueError:\n", + " continue\n", + "\n", + " return scores" + ] + }, + { + "cell_type": "markdown", + "id": "b071ce84-1866-427f-a272-b46700f364b2", + "metadata": {}, + "source": [ + "- Let's now apply this evaluation to the whole dataset and compute the average score of each model (this takes about 1 minute per model on an M3 MacBook Air laptop)\n", + "- Note that ollama is not fully deterministic across operating systems (as of this writing) so the numbers you are getting might slightly differ from the ones shown below" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "4f700d4b-19e5-4404-afa7-b0f093024232", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Scoring entries: 100%|████████████████████████| 100/100 [01:02<00:00, 1.59it/s]\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "model 1 response\n", + "Number of scores: 100 of 100\n", + "Average score: 78.48\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Scoring entries: 100%|████████████████████████| 100/100 [01:10<00:00, 1.42it/s]" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "model 2 response\n", + "Number of scores: 99 of 100\n", + "Average score: 64.98\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "\n" + ] + } + ], + "source": [ + "from pathlib import Path\n", + "\n", + "for model in (\"model 1 response\", \"model 2 response\"):\n", + "\n", + " scores = generate_model_scores(json_data, model)\n", + " print(f\"\\n{model}\")\n", + " print(f\"Number of scores: {len(scores)} of {len(json_data)}\")\n", + " print(f\"Average score: {sum(scores)/len(scores):.2f}\\n\")\n", + "\n", + " # Optionally save the scores\n", + " save_path = Path(\"scores\") / f\"llama3-8b-{model.replace(' ', '-')}.json\"\n", + " with open(save_path, \"w\") as file:\n", + " json.dump(scores, file)" + ] + }, + { + "cell_type": "markdown", + "id": "8169d534-1fec-43c4-9550-5cb701ff7f05", + "metadata": {}, + "source": [ + "- Based on the evaluation above, we can say that the 1st model is better than the 2nd model" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.6" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/03_model-evaluation/llm-instruction-eval-openai.ipynb b/Codes/ch07/03_model-evaluation/llm-instruction-eval-openai.ipynb new file mode 100644 index 0000000..01de4fd --- /dev/null +++ b/Codes/ch07/03_model-evaluation/llm-instruction-eval-openai.ipynb @@ -0,0 +1,571 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "136a4efe-fb99-4311-8679-e0a5b6282755", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "b1910a06-e8a3-40ac-8201-ff70615b1ba4", + "metadata": { + "tags": [] + }, + "source": [ + "# Evaluating Instruction Responses Using the OpenAI API" + ] + }, + { + "cell_type": "markdown", + "id": "a128651b-f326-4232-a994-42f38b7ed520", + "metadata": {}, + "source": [ + "- This notebook uses OpenAI's GPT-4 API to evaluate responses by a instruction finetuned LLMs based on an dataset in JSON format that includes the generated model responses, for example:\n", + "\n", + "\n", + "\n", + "```python\n", + "{\n", + " \"instruction\": \"What is the atomic number of helium?\",\n", + " \"input\": \"\",\n", + " \"output\": \"The atomic number of helium is 2.\", # <-- The target given in the test set\n", + " \"model 1 response\": \"\\nThe atomic number of helium is 2.0.\", # <-- Response by an LLM\n", + " \"model 2 response\": \"\\nThe atomic number of helium is 3.\" # <-- Response by a 2nd LLM\n", + "},\n", + "```" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "267ba0d1-b884-42df-85bd-0be746fd47a5", + "metadata": {}, + "outputs": [], + "source": [ + "# pip install -r requirements-extra.txt" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "63610acc-db94-437f-8d38-e99dca0299cb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "openai version: 1.30.3\n", + "tqdm version: 4.66.2\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\"openai\", # OpenAI API\n", + " \"tqdm\", # Progress bar\n", + " ]\n", + "\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "markdown", + "id": "8bcdcb34-ac75-4f4f-9505-3ce0666c42d5", + "metadata": {}, + "source": [ + "## Test OpenAI API" + ] + }, + { + "cell_type": "markdown", + "id": "9558a522-650d-401a-84fc-9fd7b1f39da7", + "metadata": {}, + "source": [ + "- First, let's test if the OpenAI API is correctly set up\n", + "- If you don't have an account yet, you need to create one at https://platform.openai.com/\n", + "- Note that you will also have to transfer some funds to your account as the GPT-4 API is not free (see https://platform.openai.com/settings/organization/billing/overview)\n", + "- Running the experiments and creating the ~200 evaluations using the code in this notebook costs about $0.26 (26 cents) as of this writing" + ] + }, + { + "cell_type": "markdown", + "id": "89343a84-0ddc-42fc-bf50-298a342b93c0", + "metadata": {}, + "source": [ + "- First, we need to provide our OpenAI API secret key, which can be found at https://platform.openai.com/api-keys\n", + "- Make sure not to share this key with anyone\n", + "- Add this secret key (`\"sk-...\"`) to the `config.json` file in this folder" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "65b0ba76-1fb1-4306-a7c2-8f3bb637ccdb", + "metadata": {}, + "outputs": [], + "source": [ + "import json\n", + "from openai import OpenAI\n", + "\n", + "# Load API key from a JSON file.\n", + "# Make sure to replace \"sk-...\" with your actual API key from https://platform.openai.com/api-keys\n", + "with open(\"config.json\", \"r\") as config_file:\n", + " config = json.load(config_file)\n", + " api_key = config[\"OPENAI_API_KEY\"]\n", + "\n", + "client = OpenAI(api_key=api_key)" + ] + }, + { + "cell_type": "markdown", + "id": "16642a48-1cab-40d2-af08-ab8c2fbf5876", + "metadata": {}, + "source": [ + "- First, let's try the API with a simple example to make sure it works as intended:" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "08e9ef2e-e816-4283-840e-43625791ad33", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'hello world'" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "def run_chatgpt(prompt, client, model=\"gpt-4-turbo\"):\n", + " response = client.chat.completions.create(\n", + " model=model,\n", + " messages=[{\"role\": \"user\", \"content\": prompt}],\n", + " temperature=0.0,\n", + " seed=123,\n", + " )\n", + " return response.choices[0].message.content\n", + "\n", + "\n", + "prompt = f\"Respond with 'hello world' if you got this message.\"\n", + "run_chatgpt(prompt, client)" + ] + }, + { + "cell_type": "markdown", + "id": "162a4739-6f03-4092-a5c2-f57a0b6a4c4d", + "metadata": {}, + "source": [ + "## Load JSON Entries" + ] + }, + { + "cell_type": "markdown", + "id": "ca011a8b-20c5-4101-979e-9b5fccf62f8a", + "metadata": {}, + "source": [ + "- Here, we assume that we saved the test dataset and the model responses as a JSON file that we can load as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "8b2d393a-aa92-4190-9d44-44326a6f699b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of entries: 100\n" + ] + } + ], + "source": [ + "json_file = \"eval-example-data.json\"\n", + "\n", + "with open(json_file, \"r\") as file:\n", + " json_data = json.load(file)\n", + "\n", + "print(\"Number of entries:\", len(json_data))" + ] + }, + { + "cell_type": "markdown", + "id": "b6c9751b-59b7-43fe-acc7-14e8daf2fa66", + "metadata": {}, + "source": [ + "- The structure of this file is as follows, where we have the given response in the test dataset (`'output'`) and responses by two different models (`'model 1 response'` and `'model 2 response'`):" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "7222fdc0-5684-4f2b-b741-3e341851359e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'instruction': 'Calculate the hypotenuse of a right triangle with legs of 6 cm and 8 cm.',\n", + " 'input': '',\n", + " 'output': 'The hypotenuse of the triangle is 10 cm.',\n", + " 'model 1 response': '\\nThe hypotenuse of the triangle is 3 cm.',\n", + " 'model 2 response': '\\nThe hypotenuse of the triangle is 12 cm.'}" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "json_data[0]" + ] + }, + { + "cell_type": "markdown", + "id": "fcf0331b-6024-4bba-89a9-a088b14a1046", + "metadata": {}, + "source": [ + "- Below is a small utility function that formats the input for visualization purposes later:" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "43263cd3-e5fb-4ab5-871e-3ad6e7d21a8c", + "metadata": {}, + "outputs": [], + "source": [ + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"Below is an instruction that describes a task. Write a response that \"\n", + " f\"appropriately completes the request.\"\n", + " f\"\\n\\n### Instruction:\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n\\n### Input:\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + " instruction_text + input_text\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "markdown", + "id": "39a55283-7d51-4136-ba60-f799d49f4098", + "metadata": {}, + "source": [ + "- Now, let's try the OpenAI API to compare the model responses (we only evaluate the first 5 responses for a visual comparison):" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "735cc089-d127-480a-b39d-0782581f0c41", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Dataset response:\n", + ">> The hypotenuse of the triangle is 10 cm.\n", + "\n", + "Model response:\n", + ">> \n", + "The hypotenuse of the triangle is 3 cm.\n", + "\n", + "Score:\n", + ">> The model response \"The hypotenuse of the triangle is 3 cm.\" is incorrect. The correct calculation of the hypotenuse for a right triangle with legs of 6 cm and 8 cm can be found using the Pythagorean theorem, which states that the square of the hypotenuse (c) is equal to the sum of the squares of the other two sides (a and b). Mathematically, this is expressed as:\n", + "\n", + "\\[ c = \\sqrt{a^2 + b^2} \\]\n", + "\\[ c = \\sqrt{6^2 + 8^2} \\]\n", + "\\[ c = \\sqrt{36 + 64} \\]\n", + "\\[ c = \\sqrt{100} \\]\n", + "\\[ c = 10 \\text{ cm} \\]\n", + "\n", + "The correct answer should be 10 cm. The response given as 3 cm is not only incorrect but also significantly off from the correct value. This error could lead to misunderstandings or incorrect applications in practical scenarios where precise measurements are crucial.\n", + "\n", + "Given the scale from 0 to 100, where 100 is the best score, the response would score very low due to its inaccuracy. However, since the response format is correct (stating the measurement and unit), it does not score the absolute minimum.\n", + "\n", + "**Score: 10/100**\n", + "\n", + "This score reflects that while the format of the response is correct, the content is highly inaccurate.\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> 1. Squirrel\n", + "2. Eagle\n", + "3. Tiger\n", + "\n", + "Model response:\n", + ">> \n", + "1. Squirrel\n", + "2. Tiger\n", + "3. Eagle\n", + "4. Cobra\n", + "5. Tiger\n", + "6. Cobra\n", + "\n", + "Score:\n", + ">> The model response lists six animals, three of which (squirrel, tiger, eagle) are indeed active during the day, making them correct responses to the instruction. However, the instruction specifically asked for three different animals, and the model response includes repetitions (tiger and cobra are each listed twice) and also exceeds the requested number of animals.\n", + "\n", + "The inclusion of \"cobra\" is incorrect as most cobras are not diurnal (active during the day); they are generally more active during the early morning and late evening, which can be considered crepuscular rather than diurnal.\n", + "\n", + "### Scoring Breakdown:\n", + "- **Relevance to the task**: The response correctly identifies three diurnal animals but also includes additional animals, which was not requested.\n", + "- **Accuracy**: Including animals not active during the day (cobra) and repeating animals reduces the accuracy.\n", + "- **Adherence to instructions**: The task was to name three different animals, but the response included six names with repetitions.\n", + "\n", + "Given these points, the response partially meets the requirements but also deviates significantly in terms of the number of animals and the inclusion of incorrect and repeated entries.\n", + "\n", + "### Score: 50/100\n", + "This score reflects that while the response did include three correct animals, it failed to strictly follow the instructions by listing only three different animals and included incorrect information.\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> I must ascertain what is incorrect.\n", + "\n", + "Model response:\n", + ">> \n", + "What is incorrect?\n", + "\n", + "Score:\n", + ">> The model response \"What is incorrect?\" scores low in terms of fulfilling the instruction to rewrite the sentence in a more formal way. The original sentence \"I need to find out what's wrong.\" expresses a personal obligation and a process of discovery, which is not captured in the model response. The model response turns the sentence into a direct question and loses the nuance of needing to discover or investigate the issue.\n", + "\n", + "**Score: 20/100**\n", + "\n", + "**Reasoning:**\n", + "- **Formality:** The response is slightly more formal than casual speech but does not elevate the formality significantly or appropriately. It does use \"incorrect\" which is slightly more formal than \"wrong.\"\n", + "- **Completeness:** The response fails to include the aspect of needing to find out or ascertain, which is a critical part of the original sentence.\n", + "- **Accuracy:** The response changes the structure and intent by converting it into a direct question, which does not align with the instruction to rewrite the statement while maintaining its original intent.\n", + "\n", + "Overall, the response does not adequately meet the requirements of the task as it significantly alters the meaning and omits key elements of the original sentence.\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> The interjection in the sentence is 'Wow'.\n", + "\n", + "Model response:\n", + ">> \n", + "The interjection in the sentence is 'Wow'.\n", + "\n", + "Score:\n", + ">> The model response `The interjection in the sentence is 'Wow'.` accurately identifies the interjection in the provided sentence. The response is clear, directly addresses the instruction, and correctly identifies \"Wow\" as the interjection, which is used to express surprise or admiration, fitting the context of the sentence. Therefore, the response is fully correct and meets all the requirements of the task.\n", + "\n", + "Score: 100/100\n", + "\n", + "-------------------------\n", + "\n", + "Dataset response:\n", + ">> The type of sentence is interrogative.\n", + "\n", + "Model response:\n", + ">> \n", + "The type of sentence is exclamatory.\n", + "\n", + "Score:\n", + ">> The model response \"The type of sentence is exclamatory.\" is incorrect. The input sentence \"Did you finish the report?\" is clearly an interrogative sentence as it is asking a question, indicated by the question mark at the end and the structure of the sentence.\n", + "\n", + "Given the scoring criteria where 100 is the best score and should be awarded to a correct and precise response, the model's response should receive a low score because it incorrectly identifies the type of sentence. An exclamatory sentence typically expresses strong emotion and ends with an exclamation mark, which is not the case here.\n", + "\n", + "Therefore, the score for the model response would be 0 out of 100, as it completely misidentifies the type of sentence, providing incorrect information.\n", + "\n", + "-------------------------\n" + ] + } + ], + "source": [ + "for entry in json_data[:5]:\n", + " prompt = (f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"score the model response `{entry['model 1 response']}`\"\n", + " f\" on a scale from 0 to 100, where 100 is the best score. \"\n", + " )\n", + " print(\"\\nDataset response:\")\n", + " print(\">>\", entry['output'])\n", + " print(\"\\nModel response:\")\n", + " print(\">>\", entry[\"model 1 response\"])\n", + " print(\"\\nScore:\")\n", + " print(\">>\", run_chatgpt(prompt, client))\n", + " print(\"\\n-------------------------\")" + ] + }, + { + "cell_type": "markdown", + "id": "142dfaa7-429f-4eb0-b74d-ff327f79547a", + "metadata": {}, + "source": [ + "- Note that the responses are very verbose; to quantify which model is better, we only want to return the scores:" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "3552bdfb-7511-42ac-a9ec-da672e2a5468", + "metadata": {}, + "outputs": [], + "source": [ + "from tqdm import tqdm\n", + "\n", + "\n", + "def generate_model_scores(json_data, json_key, client):\n", + " scores = []\n", + " for entry in tqdm(json_data, desc=\"Scoring entries\"):\n", + " prompt = (\n", + " f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"score the model response `{entry[json_key]}`\"\n", + " f\" on a scale from 0 to 100, where 100 is the best score. \"\n", + " f\"Respond with the number only.\"\n", + " )\n", + " score = run_chatgpt(prompt, client)\n", + " try:\n", + " scores.append(int(score))\n", + " except ValueError:\n", + " continue\n", + "\n", + " return scores" + ] + }, + { + "cell_type": "markdown", + "id": "71974dea-31ed-49af-abba-5c858bbbf49c", + "metadata": {}, + "source": [ + "- Please note that the response scores may vary because OpenAI's GPT models are not deterministic despite setting a random number seed, etc." + ] + }, + { + "cell_type": "markdown", + "id": "b071ce84-1866-427f-a272-b46700f364b2", + "metadata": {}, + "source": [ + "- Let's now apply this evaluation to the whole dataset and compute the average score of each model:" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "4f700d4b-19e5-4404-afa7-b0f093024232", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Scoring entries: 100%|████████████████████████| 100/100 [01:03<00:00, 1.56it/s]\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "model 1 response\n", + "Number of scores: 100 of 100\n", + "Average score: 74.09\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Scoring entries: 100%|████████████████████████| 100/100 [01:06<00:00, 1.50it/s]" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "model 2 response\n", + "Number of scores: 100 of 100\n", + "Average score: 56.57\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "\n" + ] + } + ], + "source": [ + "from pathlib import Path\n", + "\n", + "for model in (\"model 1 response\", \"model 2 response\"):\n", + "\n", + " scores = generate_model_scores(json_data, model, client)\n", + " print(f\"\\n{model}\")\n", + " print(f\"Number of scores: {len(scores)} of {len(json_data)}\")\n", + " print(f\"Average score: {sum(scores)/len(scores):.2f}\\n\")\n", + "\n", + " # Optionally save the scores\n", + " save_path = Path(\"scores\") / f\"gpt4-{model.replace(' ', '-')}.json\"\n", + " with open(save_path, \"w\") as file:\n", + " json.dump(scores, file)" + ] + }, + { + "cell_type": "markdown", + "id": "8169d534-1fec-43c4-9550-5cb701ff7f05", + "metadata": {}, + "source": [ + "- Based on the evaluation above, we can say that the 1st model is substantially better than the 2nd model" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.11" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/03_model-evaluation/requirements-extra.txt b/Codes/ch07/03_model-evaluation/requirements-extra.txt new file mode 100644 index 0000000..6446d2c --- /dev/null +++ b/Codes/ch07/03_model-evaluation/requirements-extra.txt @@ -0,0 +1,2 @@ +openai>=1.30.3 +tqdm>=4.65.0 diff --git a/Codes/ch07/03_model-evaluation/scores/correlation-analysis.ipynb b/Codes/ch07/03_model-evaluation/scores/correlation-analysis.ipynb new file mode 100644 index 0000000..edd0fa1 --- /dev/null +++ b/Codes/ch07/03_model-evaluation/scores/correlation-analysis.ipynb @@ -0,0 +1,269 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "a9bc1c1a-53bc-4b86-9140-4f1af0128037", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "5250207d-f811-46df-9d16-4ac1e9ce1c66", + "metadata": {}, + "source": [ + "# Score Correlation Analysis" + ] + }, + { + "cell_type": "markdown", + "id": "badc7ffb-d51c-4de0-97c5-b54cf3e28315", + "metadata": {}, + "source": [ + "- This notebook analyses the correlation between the different evaluation method scores" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "fa39424b-e058-4351-94ec-249b812ae8fd", + "metadata": {}, + "outputs": [], + "source": [ + "import json\n", + "\n", + "with open(\"gpt4-model-1-response.json\", \"r\") as file:\n", + " gpt4_model_1 = json.load(file)\n", + "\n", + "with open(\"llama3-8b-model-1-response.json\", \"r\") as file:\n", + " llama3_8b_model_1 = json.load(file)" + ] + }, + { + "cell_type": "markdown", + "id": "4ef67d30-7602-4695-a190-16209a152621", + "metadata": {}, + "source": [ + "## GPT-4 vs Llama 3 8B" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "2a0d4288-507f-414c-afde-9742935cd8bc", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "\n", + "\n", + "list1, list2 = gpt4_model_1, llama3_8b_model_1\n", + "\n", + "plt.scatter(list1, list2)\n", + "plt.plot(\n", + " np.unique(list1),\n", + " np.poly1d(np.polyfit(list1, list2, 1))(np.unique(list1))\n", + ")\n", + "plt.xlabel(\"GPT-4\")\n", + "plt.ylabel(\"Llama3 8B\")\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "id": "3e28b2d2-7f31-4c5f-853b-1e71dc715a25", + "metadata": {}, + "source": [ + "### Correlation Coefficients" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "44ef7e9a-1f07-4e94-bdc5-d5271616ef6d", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
PearsonSpearmanKendall Tau
Results0.804890.6984060.57292
\n", + "
" + ], + "text/plain": [ + " Pearson Spearman Kendall Tau\n", + "Results 0.80489 0.698406 0.57292" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "import pandas as pd\n", + "from scipy.stats import spearmanr, kendalltau\n", + "\n", + "pearson_correlation = np.corrcoef(list1, list2)[0, 1]\n", + "spearman_correlation, _ = spearmanr(list1, list2)\n", + "kendall_tau_correlation, _ = kendalltau(list1, list2)\n", + "\n", + "correlation_table = pd.DataFrame({\n", + " \"Pearson\": [pearson_correlation],\n", + " \"Spearman\": [spearman_correlation],\n", + " \"Kendall Tau\": [kendall_tau_correlation]\n", + "}, index=['Results'])\n", + "\n", + "correlation_table" + ] + }, + { + "cell_type": "markdown", + "id": "3a1bd708-ba5d-4290-abe3-ee736059c2cd", + "metadata": {}, + "source": [ + "- For comparison, below are the correlation coefficients from the Prometheus 2 paper by Kim et al. 2024 ([https://arxiv.org/abs/2405.01535](https://arxiv.org/abs/2405.01535)), which are all in the same ballpark as the ones reported for Llama 3 above\n", + "- Note that Prometheus 2 is a model specifically finetuned for LLM rating and evaluation " + ] + }, + { + "cell_type": "markdown", + "id": "fbc033f4-8a11-42be-a683-6cef7eb23468", + "metadata": {}, + "source": [ + "#### Pearson\n", + "\n", + "| Evaluator LM | VICUNA Bench | VICUNA Bench | MT Bench | MT Bench | FLASK | FLASK | FLASK | Feedback Bench |\n", + "|-----------------------|--------------|---------------|------------|---------------|------------|---------------|-----------|----------------|\n", + "| | GPT-4-1106 | Claude-3-Opus | GPT-4-1106 | Claude-3-Opus | GPT-4-1106 | Claude-3-Opus | Humans | GPT-4-0613 |\n", + "| LLAMA2-CHAT 7B | 0.205 | 0.243 | 0.036 | 0.055 | 0.317 | 0.256 | 0.299 | 0.523 |\n", + "| LLAMA2-CHAT 13B | 0.185 | 0.141 | -0.042 | -0.002 | 0.239 | 0.247 | 0.263 | 0.545 |\n", + "| LLAMA2-CHAT 70B | 0.350 | 0.463 | 0.178 | 0.228 | 0.388 | 0.402 | 0.317 | 0.592 |\n", + "| MISTRAL-INSTRUCT-7B | 0.486 | 0.561 | 0.284 | 0.396 | 0.448 | 0.437 | 0.377 | 0.586 |\n", + "| MIXTRAL-INSTRUCT-8X7B | 0.566 | 0.579 | 0.551 | 0.539 | 0.483 | 0.495 | 0.420 | 0.673 |\n", + "| **PROMETHEUS-7B** | **0.484** | **0.528** | **0.378** | **0.382** | **0.352** | **0.331** | **0.348** | **0.847** |\n", + "| **PROMETHEUS-13B** | **0.492** | **0.534** | **0.404** | **0.477** | **0.462** | **0.470** | **0.449** | **0.860** |\n", + "| AUTO-J (13B) | 0.351 | 0.262 | 0.432 | 0.375 | 0.430 | 0.370 | 0.473 | 0.637 |\n", + "| **PROMETHEUS-2-7B** | **0.642** | **0.610** | **0.543** | **0.554** | **0.645** | **0.578** | **0.544** | **0.878** |\n", + "| **PROMETHEUS-2-8X7B** | **0.685** | **0.635** | **0.665** | **0.614** | **0.659** | **0.626** | **0.555** | **0.898** |\n", + "| GPT-3.5-TURBO-0613 | 0.335 | 0.349 | 0.183 | 0.194 | 0.437 | 0.396 | 0.450 | 0.594 |\n", + "| GPT-4-1106 | / | 0.694 | / | 0.717 | / | 0.736 | 0.679 | 0.753 |\n", + "| CLAUDE-3-OPUS | 0.694 | / | 0.717 | / | 0.736 | / | 0.573 | 0.788 |\n", + "\n", + "#### Spearman\n", + "\n", + "| Evaluator LM | VICUNA Bench | VICUNA Bench | MT Bench | MT Bench | MT Bench | FLASK | FLASK | Feedback Bench |\n", + "|-----------------------|--------------|---------------|------------|---------------|------------|---------------|-----------|----------------|\n", + "| | GPT-4-1106 | Claude-3-Opus | GPT-4-1106 | Claude-3-Opus | GPT-4-1106 | Claude-3-Opus | Humans | GPT-4-0613 |\n", + "| LLAMA2-CHAT 7B | 0.236 | 0.255 | 0.084 | 0.089 | 0.301 | 0.244 | 0.279 | 0.511 |\n", + "| LLAMA2-CHAT 13B | 0.178 | 0.179 | -0.025 | 0.044 | 0.206 | 0.222 | 0.224 | 0.543 |\n", + "| LLAMA2-CHAT 70B | 0.348 | 0.466 | 0.197 | 0.252 | 0.391 | 0.389 | 0.298 | 0.585 |\n", + "| MISTRAL-INSTRUCT-7B | 0.389 | 0.480 | 0.266 | 0.358 | 0.499 | 0.478 | 0.374 | 0.563 |\n", + "| MIXTRAL-INSTRUCT-8X7B | 0.476 | 0.556 | 0.545 | 0.517 | 0.505 | 0.500 | 0.386 | 0.659 |\n", + "| **PROMETHEUS-7B** | **0.508** | **0.528** | **0.385** | **0.349** | **0.367** | **0.326** | **0.317** | **0.876** |\n", + "| **PROMETHEUS-13B** | **0.492** | **0.534** | **0.401** | **0.470** | **0.474** | **0.454** | **0.398** | **0.893** |\n", + "| AUTO-J (13B) | 0.337 | 0.297 | 0.408 | 0.365 | 0.402 | 0.358 | 0.408 | 0.623 |\n", + "| **PROMETHEUS-2-7B** | **0.643** | **0.584** | **0.550** | **0.524** | **0.626** | **0.569** | **0.490** | **0.909** |\n", + "| **PROMETHEUS-2-8X7B** | **0.660** | **0.615** | **0.669** | **0.605** | **0.642** | **0.618** | **0.496** | **0.912** |\n", + "| GPT-3.5-TURBO-0613 | 0.319 | 0.354 | 0.192 | 0.198 | 0.446 | 0.390 | 0.374 | 0.565 |\n", + "| GPT-4-1106 | / | 0.659 | / | 0.721 | / | 0.729 | 0.650 | 0.753 |\n", + "| CLAUDE-3-OPUS | 0.659 | / | 0.721 | / | 0.729 | / | 0.567 | 0.784 |\n", + "\n", + "#### Kendall-Tau\n", + "\n", + "| Evaluator LM | VICUNA Bench | VICUNA Bench | MT Bench | MT Bench | FLASK | FLASK | FLASK | Feedback Bench |\n", + "|-----------------------|--------------|---------------|------------|---------------|------------|---------------|-----------|----------------|\n", + "| | GPT-4-1106 | Claude-3-Opus | GPT-4-1106 | Claude-3-Opus | GPT-4-1106 | Claude-3-Opus | Humans | GPT-4-0613 |\n", + "| LLAMA2-CHAT 7B | 0.183 | 0.203 | 0.065 | 0.070 | 0.229 | 0.186 | 0.211 | 0.419 |\n", + "| LLAMA2-CHAT 13B | 0.145 | 0.146 | -0.019 | 0.037 | 0.160 | 0.174 | 0.174 | 0.453 |\n", + "| LLAMA2-CHAT 70B | 0.282 | 0.382 | 0.150 | 0.196 | 0.310 | 0.310 | 0.221 | 0.487 |\n", + "| MISTRAL-INSTRUCT-7B | 0.314 | 0.391 | 0.208 | 0.281 | 0.395 | 0.384 | 0.287 | 0.454 |\n", + "| MIXTRAL-INSTRUCT-8X7B | 0.395 | 0.468 | 0.433 | 0.419 | 0.410 | 0.408 | 0.304 | 0.551 |\n", + "| **PROMETHEUS-7B** | **0.405** | **0.425** | **0.290** | **0.263** | **0.282** | **0.251** | **0.236** | **0.770** |\n", + "| **PROMETHEUS-13B** | **0.397** | **0.434** | **0.299** | **0.352** | **0.365** | **0.352** | **0.299** | **0.793** |\n", + "| AUTO-J (13B) | 0.282 | 0.242 | 0.303 | 0.272 | 0.312 | 0.282 | 0.312 | 0.515 |\n", + "| **PROMETHEUS-2-7B** | **0.515** | **0.478** | **0.458** | **0.421** | **0.500** | **0.454** | **0.376** | **0.773** |\n", + "| **PROMETHEUS-2-8X7B** | **0.559** | **0.515** | **0.535** | **0.483** | **0.526** | **0.507** | **0.388** | **0.800** |\n", + "| GPT-3.5-TURBO-0613 | 0.255 | 0.287 | 0.148 | 0.157 | 0.360 | 0.315 | 0.298 | 0.489 |\n", + "| GPT-4-1106 | / | 0.553 | / | 0.590 | / | 0.609 | 0.517 | 0.662 |\n", + "| CLAUDE-3-OPUS | 0.553 | / | 0.590 | / | 0.609 | / | 0.453 | 0.693 |" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.4" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/03_model-evaluation/scores/gpt4-model-1-response.json b/Codes/ch07/03_model-evaluation/scores/gpt4-model-1-response.json new file mode 100644 index 0000000..4afd2ca --- /dev/null +++ b/Codes/ch07/03_model-evaluation/scores/gpt4-model-1-response.json @@ -0,0 +1 @@ +[0, 50, 20, 100, 0, 100, 0, 100, 100, 100, 55, 0, 100, 100, 100, 100, 100, 0, 98, 100, 100, 0, 100, 100, 100, 100, 100, 100, 0, 100, 100, 0, 100, 100, 85, 100, 0, 0, 100, 100, 100, 100, 100, 100, 0, 100, 100, 95, 20, 50, 85, 100, 100, 100, 100, 55, 100, 100, 100, 0, 100, 98, 100, 100, 100, 0, 85, 100, 100, 98, 100, 100, 100, 0, 100, 100, 100, 100, 0, 100, 0, 100, 100, 0, 0, 100, 50, 100, 100, 10, 100, 100, 100, 100, 0, 100, 100, 25, 100, 30] \ No newline at end of file diff --git a/Codes/ch07/03_model-evaluation/scores/gpt4-model-2-response.json b/Codes/ch07/03_model-evaluation/scores/gpt4-model-2-response.json new file mode 100644 index 0000000..6daf9cc --- /dev/null +++ b/Codes/ch07/03_model-evaluation/scores/gpt4-model-2-response.json @@ -0,0 +1 @@ +[0, 100, 0, 100, 0, 100, 0, 100, 0, 0, 50, 0, 100, 100, 100, 100, 100, 100, 100, 95, 0, 50, 100, 100, 0, 0, 100, 0, 0, 100, 0, 0, 100, 0, 67, 0, 0, 0, 100, 100, 95, 100, 100, 100, 0, 0, 0, 0, 100, 100, 100, 0, 55, 100, 0, 100, 65, 100, 100, 0, 100, 100, 100, 0, 100, 0, 85, 100, 100, 85, 0, 75, 100, 0, 0, 100, 100, 100, 0, 100, 0, 50, 100, 100, 0, 100, 0, 0, 100, 85, 100, 0, 100, 100, 0, 100, 100, 0, 0, 0] \ No newline at end of file diff --git a/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-1-response.json b/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-1-response.json new file mode 100644 index 0000000..d991d71 --- /dev/null +++ b/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-1-response.json @@ -0,0 +1 @@ +[20, 92, 85, 90, 20, 90, 22, 97, 60, 96, 20, 20, 98, 95, 90, 98, 95, 20, 98, 98, 92, 20, 96, 96, 100, 98, 98, 95, 20, 95, 98, 20, 85, 95, 80, 97, 40, 21, 100, 85, 95, 98, 92, 98, 69, 98, 80, 60, 60, 20, 80, 68, 80, 96, 96, 68, 80, 95, 80, 20, 95, 98, 80, 98, 94, 20, 40, 98, 100, 85, 98, 90, 95, 85, 95, 80, 98, 98, 25, 98, 40, 92, 95, 82, 87, 98, 80, 90, 95, 4, 90, 90, 80, 98, 20, 98, 98, 40, 92, 98] \ No newline at end of file diff --git a/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-2-response.json b/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-2-response.json new file mode 100644 index 0000000..b834c71 --- /dev/null +++ b/Codes/ch07/03_model-evaluation/scores/llama3-8b-model-2-response.json @@ -0,0 +1 @@ +[76, 85, 67, 90, 20, 98, 22, 96, 40, 80, 40, 20, 90, 98, 80, 92, 98, 98, 95, 99, 55, 99, 80, 90, 20, 4, 98, 4, 40, 95, 14, 44, 95, 44, 80, 4, 4, 40, 95, 80, 98, 95, 92, 98, 68, 20, 20, 60, 95, 90, 98, 0, 20, 80, 20, 80, 92, 98, 98, 20, 95, 100, 95, 85, 98, 4, 40, 98, 98, 65, 20, 76, 100, 67, 44, 92, 75, 97, 27, 98, 20, 60, 90, 96, 67, 98, 80, 10, 80, 98, 100, 40, 92, 98, 20, 98, 98, 20, 20] \ No newline at end of file diff --git a/Codes/ch07/04_preference-tuning-with-dpo/README.md b/Codes/ch07/04_preference-tuning-with-dpo/README.md new file mode 100644 index 0000000..3b71a64 --- /dev/null +++ b/Codes/ch07/04_preference-tuning-with-dpo/README.md @@ -0,0 +1,7 @@ +# Chapter 7: Finetuning to Follow Instructions + +- [create-preference-data-ollama.ipynb](create-preference-data-ollama.ipynb): A notebook that creates a synthetic dataset for preference finetuning dataset using Llama 3.1 and Ollama + +- [dpo-from-scratch.ipynb](dpo-from-scratch.ipynb): This notebook implements Direct Preference Optimization (DPO) for LLM alignment + + diff --git a/Codes/ch07/04_preference-tuning-with-dpo/create-preference-data-ollama.ipynb b/Codes/ch07/04_preference-tuning-with-dpo/create-preference-data-ollama.ipynb new file mode 100644 index 0000000..41c948e --- /dev/null +++ b/Codes/ch07/04_preference-tuning-with-dpo/create-preference-data-ollama.ipynb @@ -0,0 +1,595 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "136a4efe-fb99-4311-8679-e0a5b6282755", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "b1910a06-e8a3-40ac-8201-ff70615b1ba4", + "metadata": { + "tags": [] + }, + "source": [ + "# Generating A Preference Dataset With Llama 3.1 70B And Ollama" + ] + }, + { + "cell_type": "markdown", + "id": "a128651b-f326-4232-a994-42f38b7ed520", + "metadata": {}, + "source": [ + "- Preference finetuning is a process to align an instruction-finetuned LLM with human preferences\n", + "- There are multiple ways to create a dataset for preference finetuning an LLM\n", + " 1. We use the instruction-finetuned LLM to generate multiple responses and have humans rank them based on their preference and/or given preference criteria\n", + " 2. We use the instruction-finetuned LLM to generate multiple responses and have LLMs rank them based on given preference criteria\n", + " 3. We use an LLM to generate preferred and dispreferred responses given certain preference criteria\n", + "- In this notebook, we consider approach 3\n", + "- This notebook uses a 70-billion-parameter Llama 3.1-Instruct model through ollama to generate preference labels for an instruction dataset\n", + "- The expected format of the instruction dataset is as follows:\n", + "\n", + "\n", + "### Input\n", + "\n", + "```json\n", + "[\n", + " {\n", + " \"instruction\": \"What is the state capital of California?\",\n", + " \"input\": \"\",\n", + " \"output\": \"The state capital of California is Sacramento.\",\n", + " },\n", + " {\n", + " \"instruction\": \"Provide a synonym for 'fast'.\",\n", + " \"input\": \"\",\n", + " \"output\": \"A synonym for 'fast' is 'quick'.\",\n", + " },\n", + " {\n", + " \"instruction\": \"What is the capital of Greece?\",\n", + " \"input\": \"\",\n", + " \"output\": \"The capital of Greece is Athens.\",\n", + "\n", + " },\n", + "...\n", + "]\n", + "```\n", + "\n", + "The output dataset will look as follows, where more polite responses are preferred (`'chosen'`), and more impolite responses are dispreferred (`'rejected'`):\n", + "\n", + "```json\n", + "[\n", + " {\n", + " \"instruction\": \"What is the state capital of California?\",\n", + " \"input\": \"\",\n", + " \"output\": \"The state capital of California is Sacramento.\",\n", + " \"rejected\": \"Look, the state capital of California is obviously Sacramento.\",\n", + " \"chosen\": \"The state capital of California is Sacramento.\"\n", + " },\n", + " {\n", + " \"instruction\": \"Provide a synonym for 'fast'.\",\n", + " \"input\": \"\",\n", + " \"output\": \"A synonym for 'fast' is 'quick'.\",\n", + " \"chosen\": \"A suitable alternative to 'fast' would be 'quick'.\",\n", + " \"rejected\": \"A synonym for 'fast' is 'quick'.\"\n", + " },\n", + " {\n", + " \"instruction\": \"What is the capital of Greece?\",\n", + " \"input\": \"\",\n", + " \"output\": \"The capital of Greece is Athens.\",\n", + " \"chosen\": \"I'd be happy to help! The capital of Greece is indeed Athens.\",\n", + " \"rejected\": \"The capital of Greece is Athens.\"\n", + " },\n", + "...\n", + "]\n", + "```\n", + "\n", + "### Output\n", + "\n", + "\n", + "\n", + "\n", + "- The code doesn't require a GPU and runs on a laptop given enough RAM" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "63610acc-db94-437f-8d38-e99dca0299cb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tqdm version: 4.66.4\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\"tqdm\", # Progress bar\n", + " ]\n", + "\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "markdown", + "id": "8bcdcb34-ac75-4f4f-9505-3ce0666c42d5", + "metadata": {}, + "source": [ + "## Installing Ollama and Downloading Llama 3.1" + ] + }, + { + "cell_type": "markdown", + "id": "5a092280-5462-4709-a3fe-8669a4a8a0a6", + "metadata": {}, + "source": [ + "- Ollama is an application to run LLMs efficiently\n", + "- It is a wrapper around [llama.cpp](https://github.com/ggerganov/llama.cpp), which implements LLMs in pure C/C++ to maximize efficiency\n", + "- Note that it is a tool for using LLMs to generate text (inference), not training or finetuning LLMs\n", + "- Prior to running the code below, install ollama by visiting [https://ollama.com](https://ollama.com) and following the instructions (for instance, clicking on the \"Download\" button and downloading the ollama application for your operating system)" + ] + }, + { + "cell_type": "markdown", + "id": "9558a522-650d-401a-84fc-9fd7b1f39da7", + "metadata": {}, + "source": [ + "- For macOS and Windows users, click on the ollama application you downloaded; if it prompts you to install the command line usage, say \"yes\"\n", + "- Linux users can use the installation command provided on the ollama website\n", + "\n", + "- In general, before we can use ollama from the command line, we have to either start the ollama application or run `ollama serve` in a separate terminal\n", + "\n", + "\n", + "\n", + "\n", + "- With the ollama application or `ollama serve` running, in a different terminal, on the command line, execute the following command to try out the 70-billion-parameter Llama 3.1 model \n", + "\n", + "```bash\n", + "# 70B model\n", + "ollama run llama3.1:70b\n", + "```\n", + "\n", + "\n", + "The output looks like as follows:\n", + "\n", + "```\n", + "$ ollama run llama3.1:70b\n", + "pulling manifest\n", + "pulling aa81b541aae6... 100% ▕████████████████▏ 39 GB\n", + "pulling 8cf247399e57... 100% ▕████████████████▏ 1.7 KB\n", + "pulling f1cd752815fc... 100% ▕████████████████▏ 12 KB\n", + "pulling 56bb8bd477a5... 100% ▕████████████████▏ 96 B\n", + "pulling 3c1c2d3df5b3... 100% ▕████████████████▏ 486 B\n", + "verifying sha256 digest\n", + "writing manifest\n", + "removing any unused layers\n", + "success\n", + "```\n", + "\n", + "- Note that `llama3.1:70b` refers to the instruction finetuned 70-billion-parameter Llama 3.1 model\n", + "\n", + "- Alternatively, you can also use the smaller, more resource-effiicent 8-billion-parameters Llama 3.1 model, by replacing `llama3.1:70b` with `llama3.1`\n", + "\n", + "- After the download has been completed, you will see a command line prompt that allows you to chat with the model\n", + "\n", + "- Try a prompt like \"What do llamas eat?\", which should return an output similar to the following:\n", + "\n", + "```\n", + ">>> What do llamas eat?\n", + "Llamas are ruminant animals, which means they have a four-chambered \n", + "stomach and eat plants that are high in fiber. In the wild, llamas \n", + "typically feed on:\n", + "1. Grasses: They love to graze on various types of grasses, including tall \n", + "grasses, wheat, oats, and barley.\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "0b5addcb-fc7d-455d-bee9-6cc7a0d684c7", + "metadata": {}, + "source": [ + "- You can end this session using the input `/bye`" + ] + }, + { + "cell_type": "markdown", + "id": "dda155ee-cf36-44d3-b634-20ba8e1ca38a", + "metadata": {}, + "source": [ + "## Using Ollama's REST API" + ] + }, + { + "cell_type": "markdown", + "id": "89343a84-0ddc-42fc-bf50-298a342b93c0", + "metadata": {}, + "source": [ + "- Now, an alternative way to interact with the model is via its REST API in Python via the following function\n", + "- Before you run the next cells in this notebook, make sure that ollama is still running, as described above, via\n", + " - `ollama serve` in a terminal\n", + " - the ollama application\n", + "- Next, run the following code cell to query the model" + ] + }, + { + "cell_type": "markdown", + "id": "16642a48-1cab-40d2-af08-ab8c2fbf5876", + "metadata": {}, + "source": [ + "- First, let's try the API with a simple example to make sure it works as intended:" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "65b0ba76-1fb1-4306-a7c2-8f3bb637ccdb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Llamas are herbivores, which means they primarily eat plants and plant-based foods. Their diet consists of:\n", + "\n", + "1. **Grasses**: Various types of grasses, including timothy grass, orchard grass, and brome grass.\n", + "2. **Hay**: High-quality hay, such as alfalfa or clover hay, is a staple in a llama's diet.\n", + "3. **Leaves**: Leaves from trees and shrubs, like willow, cottonwood, and mesquite, are also eaten.\n", + "4. **Fruits and vegetables**: Llamas enjoy fruits like apples, carrots, and sweet potatoes, as well as leafy greens like kale and spinach.\n", + "5. **Grains**: In moderation, llamas can eat grains like oats, barley, and corn.\n", + "\n", + "It's essential to note that llamas have a unique digestive system, with a three-part stomach and a large cecum (a specialized part of the large intestine). This allows them to break down and extract nutrients from plant material more efficiently than many other animals.\n", + "\n", + "A typical llama diet might consist of:\n", + "\n", + "* 1-2% of their body weight in hay per day\n", + "* 0.5-1% of their body weight in grains per day (if fed)\n", + "* Free-choice access to fresh water\n", + "* Limited amounts of fruits and vegetables as treats\n", + "\n", + "It's also important to ensure that llamas have access to a mineral supplement, such as a salt lick or loose minerals, to help maintain optimal health.\n", + "\n", + "Remember, every llama is different, and their dietary needs may vary depending on factors like age, size, and activity level. Consult with a veterinarian or experienced llama breeder for specific guidance on feeding your llama.\n" + ] + } + ], + "source": [ + "import urllib.request\n", + "import json\n", + "\n", + "\n", + "def query_model(prompt, model=\"llama3.1:70b\", url=\"http://localhost:11434/api/chat\"):\n", + " # Create the data payload as a dictionary\n", + " data = {\n", + " \"model\": model,\n", + " \"messages\": [\n", + " {\n", + " \"role\": \"user\",\n", + " \"content\": prompt\n", + " }\n", + " ],\n", + " \"options\": {\n", + " \"seed\": 123,\n", + " \"temperature\": 0,\n", + " }\n", + " }\n", + "\n", + " # Convert the dictionary to a JSON formatted string and encode it to bytes\n", + " payload = json.dumps(data).encode(\"utf-8\")\n", + "\n", + " # Create a request object, setting the method to POST and adding necessary headers\n", + " request = urllib.request.Request(url, data=payload, method=\"POST\")\n", + " request.add_header(\"Content-Type\", \"application/json\")\n", + "\n", + " # Send the request and capture the response\n", + " response_data = \"\"\n", + " with urllib.request.urlopen(request) as response:\n", + " # Read and decode the response\n", + " while True:\n", + " line = response.readline().decode(\"utf-8\")\n", + " if not line:\n", + " break\n", + " response_json = json.loads(line)\n", + " response_data += response_json[\"message\"][\"content\"]\n", + "\n", + " return response_data\n", + "\n", + "\n", + "result = query_model(\"What do Llamas eat?\")\n", + "print(result)" + ] + }, + { + "cell_type": "markdown", + "id": "162a4739-6f03-4092-a5c2-f57a0b6a4c4d", + "metadata": {}, + "source": [ + "## Load JSON Entries" + ] + }, + { + "cell_type": "markdown", + "id": "ca011a8b-20c5-4101-979e-9b5fccf62f8a", + "metadata": {}, + "source": [ + "- Now, let's get to the data generation part\n", + "- Here, for a hands-on example, we use the `instruction-data.json` file that we originally used to instruction-finetune the model in chapter 7:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "8b2d393a-aa92-4190-9d44-44326a6f699b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of entries: 1100\n" + ] + } + ], + "source": [ + "from pathlib import Path\n", + "\n", + "json_file = Path(\"..\", \"01_main-chapter-code\", \"instruction-data.json\")\n", + "\n", + "with open(json_file, \"r\") as file:\n", + " json_data = json.load(file)\n", + "\n", + "print(\"Number of entries:\", len(json_data))" + ] + }, + { + "cell_type": "markdown", + "id": "b6c9751b-59b7-43fe-acc7-14e8daf2fa66", + "metadata": {}, + "source": [ + "- The structure of this file is as follows, where we have the given response in the test dataset (`'output'`) that we trained the model to generate via instruction finetuning based on the `'input'` and `'instruction'`" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "7222fdc0-5684-4f2b-b741-3e341851359e", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'instruction': 'Evaluate the following phrase by transforming it into the spelling given.',\n", + " 'input': 'freind --> friend',\n", + " 'output': 'The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\".'}" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "json_data[0]" + ] + }, + { + "cell_type": "markdown", + "id": "fcf0331b-6024-4bba-89a9-a088b14a1046", + "metadata": {}, + "source": [ + "- Below is a small utility function that formats the instruction and input:" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "43263cd3-e5fb-4ab5-871e-3ad6e7d21a8c", + "metadata": {}, + "outputs": [], + "source": [ + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"Below is an instruction that describes a task. Write a response that \"\n", + " f\"appropriately completes the request.\"\n", + " f\"\\n\\n### Instruction:\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n\\n### Input:\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + " instruction_text + input_text\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "markdown", + "id": "39a55283-7d51-4136-ba60-f799d49f4098", + "metadata": {}, + "source": [ + "- Now, let's try the ollama API to generate a `'chosen'` and `'rejected'` response for preference tuning a model\n", + "- Here, to for illustration purposes, we create answers that are more or less polite\n" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "735cc089-d127-480a-b39d-0782581f0c41", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "Dataset response:\n", + ">> The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\".\n", + "\n", + "impolite response:\n", + ">> The spelling of the given phrase \"freind\" is flat out wrong, get it together, the correct spelling is \"friend\".\n", + "\n", + "Dataset response:\n", + ">> He goes to the park every day.\n", + "\n", + "polite response:\n", + ">> He goes to the park daily, if I'm not mistaken.\n", + "\n", + "Dataset response:\n", + ">> 45 kilometers is 45000 meters.\n", + "\n", + "polite response:\n", + ">> 45 kilometers is equivalent to 45000 meters.\n", + "\n", + "Dataset response:\n", + ">> Although it was raining, they went for a walk.\n", + "\n", + "polite response:\n", + ">> Although it was raining outside, they still decided to go for a walk.\n", + "\n", + "Dataset response:\n", + ">> 1, 4, 9, 16, 25, 36, 49, 64, 81, 100.\n", + "\n", + "impolite response:\n", + ">> Here are your precious square numbers: 1, 4, 9, 16, 25, 36, 49, 64, 81, 100.\n" + ] + } + ], + "source": [ + "import random\n", + "\n", + "\n", + "for entry in json_data[:5]:\n", + " \n", + " politeness = random.choice([\"polite\", \"impolite\"]) \n", + " prompt = (\n", + " f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"slightly rewrite the output to be more {politeness}.\"\n", + " \"Keep the modification minimal.\"\n", + " \"Only return return the generated response and nothing else.\"\n", + " )\n", + " print(\"\\nDataset response:\")\n", + " print(\">>\", entry['output'])\n", + " print(f\"\\n{politeness} response:\")\n", + " print(\">>\", query_model(prompt)) " + ] + }, + { + "cell_type": "markdown", + "id": "142dfaa7-429f-4eb0-b74d-ff327f79547a", + "metadata": {}, + "source": [ + "- If we find that the generated responses above look reasonable, we can go to the next step and apply the prompt to the whole dataset\n", + "- Here, we add a `'chosen'` key for the preferred response and a `'rejected'` response for the dispreferred response" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "3349dbbc-963f-4af3-9790-12dbfdca63c3", + "metadata": {}, + "outputs": [], + "source": [ + "import random\n", + "from tqdm import tqdm\n", + "\n", + "def generate_model_responses(json_data):\n", + "\n", + " for i, entry in enumerate(tqdm(json_data, desc=\"Writing entries\")):\n", + " politeness = random.choice([\"polite\", \"impolite\"]) \n", + " prompt = (\n", + " f\"Given the input `{format_input(entry)}` \"\n", + " f\"and correct output `{entry['output']}`, \"\n", + " f\"slightly rewrite the output to be more {politeness}.\"\n", + " \"Keep the modification minimal.\"\n", + " \"Only return return the generated response and nothing else.\"\n", + " )\n", + " response = query_model(prompt)\n", + " \n", + " if politeness == \"polite\":\n", + " json_data[i][\"chosen\"] = response\n", + " json_data[i][\"rejected\"] = entry[\"output\"]\n", + " else:\n", + " json_data[i][\"rejected\"] = response\n", + " json_data[i][\"chosen\"] = entry[\"output\"] " + ] + }, + { + "cell_type": "markdown", + "id": "b071ce84-1866-427f-a272-b46700f364b2", + "metadata": {}, + "source": [ + "- Let's now apply this evaluation to the whole dataset and compute the average score of each model (this takes about 1 minute per model on an M3 MacBook Air laptop)\n", + "- Note that ollama is not fully deterministic across operating systems (as of this writing) so the numbers you are getting might slightly differ from the ones shown below" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "4f700d4b-19e5-4404-afa7-b0f093024232", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Writing entries: 100%|██████████| 1100/1100 [17:20<00:00, 1.06it/s]\n" + ] + } + ], + "source": [ + "generate_model_responses(json_data)" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "838d9747-0f7d-46fe-aab5-9ee6b765d021", + "metadata": {}, + "outputs": [], + "source": [ + "with open(\"instruction-data-with-preference.json\", \"w\") as file:\n", + " json.dump(json_data, file, indent=4)" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.6" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb b/Codes/ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb new file mode 100644 index 0000000..c8773b1 --- /dev/null +++ b/Codes/ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb @@ -0,0 +1,3096 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "62129596-d10f-45b1-a1af-ee10f358f773", + "metadata": { + "id": "62129596-d10f-45b1-a1af-ee10f358f773" + }, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "b0bd2379-ed2f-4c77-8b71-f1f0242b9ff9", + "metadata": { + "id": "b0bd2379-ed2f-4c77-8b71-f1f0242b9ff9" + }, + "source": [ + "# Direct Preference Optimization (DPO) for LLM Alignment (From Scratch)" + ] + }, + { + "cell_type": "markdown", + "id": "d04cb2b8-d87b-4c6b-a225-c630d758f68e", + "metadata": { + "id": "d04cb2b8-d87b-4c6b-a225-c630d758f68e" + }, + "source": [ + "- This code notebook implements Direct Preference Optimization (DPO) from scratch and applies it to a large language model (LLM) to enhance its ability to generate responses that align more closely with user preferences" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "pxMGAf3bnVwn", + "metadata": { + "id": "pxMGAf3bnVwn" + }, + "outputs": [], + "source": [ + "# !pip install -r https://raw.githubusercontent.com/rasbt/LLMs-from-scratch/main/requirements.txt" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "edb3e145-fbaa-4bb3-9e95-186b4145087f", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "edb3e145-fbaa-4bb3-9e95-186b4145087f", + "outputId": "3d449525-76cc-4124-ab30-a93c6a9623ee" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tiktoken version: 0.7.0\n", + "torch version: 2.3.1+cu121\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\n", + " \"tiktoken\", # Tokenizer\n", + " \"torch\", # Deep learning library\n", + "]\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "markdown", + "id": "49ec20a3-a26c-4f9b-8a33-bfd3d67860e2", + "metadata": { + "id": "49ec20a3-a26c-4f9b-8a33-bfd3d67860e2" + }, + "source": [ + " \n", + "# 1) A brief introduction to DPO" + ] + }, + { + "cell_type": "markdown", + "id": "17804afd-786b-4600-bad0-f5805454e3d6", + "metadata": { + "id": "17804afd-786b-4600-bad0-f5805454e3d6" + }, + "source": [ + "- DPO, proposed in the paper [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://arxiv.org/abs/2305.18290), is an alternative to reinforcement learning from human feedback (RLHF) used in finetuning large language models (LLMs)\n", + "- DPO can be used to finetune (or align) the model to generate responses that better align with user expectations and instructions\n", + "\n", + "\n", + "\n", + "- In instruction finetuning, we train the LLM to generate correct answers given a prompt\n", + "- However, in practice, there are multiple ways to give a correct answer, and correct answers can differ in style; for example, consider a technical and a more user-friendly response when asking an LLM to give recommendations when buying a laptop, as shown in the figure below\n", + "\n", + "\n", + "\n", + "- RLHF and DPO are methods that can be used to teach the LLM to prefer one answer style over the other, that is, aligning better with user preferences\n", + "- The RLHF process, which requires training a separate reward model, is outlined below\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "9073622f-d537-42bf-8778-43c2adaa2191", + "metadata": { + "id": "9073622f-d537-42bf-8778-43c2adaa2191" + }, + "source": [ + "- Compared to RLHF, DPO aims to simplify the process by optimizing models directly for user preferences without the need for complex reward modeling and policy optimization\n", + "- In other words, DPO focuses on directly optimizing the model's output to align with human preferences or specific objectives\n", + "- Shown below is the main idea as an overview of how DPO works\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "c894134a-315c-453e-bbc1-387794b3f4d6", + "metadata": { + "id": "c894134a-315c-453e-bbc1-387794b3f4d6" + }, + "source": [ + "- The concrete equation to implement the DPO loss is shown below; we will revisit the equation when we implement it in Python further down in this code notebook\n", + "\n", + "" + ] + }, + { + "cell_type": "markdown", + "id": "dd7491b5-f619-4501-ad39-2942de57c115", + "metadata": { + "id": "dd7491b5-f619-4501-ad39-2942de57c115" + }, + "source": [ + "- In the equation above,\n", + " - \"expected value\" $\\mathbb{E}$ is statistics jargon and stands for the average or mean value of the random variable (the expression inside the brackets)\n", + " - The $\\pi_{\\theta}$ variable is the so-called policy (a term borrowed from reinforcement learning) and represents the LLM we want to optimize; $\\pi_{ref}$ is a reference LLM, which is typically the original LLM before optimization (at the beginning of the training, $\\pi_{\\theta}$ and $\\pi_{ref}$ are typically the same)\n", + " - $\\beta$ is a hyperparameter to control the divergence between the $\\pi_{\\theta}$ and the reference model; increasing $\\beta$ increases the impact of the difference between\n", + "$\\pi_{\\theta}$ and $\\pi_{ref}$ in terms of their log probabilities on the overall loss function, thereby increasing the divergence between the two models\n", + "- To avoid bloating the code notebook with a more detailed discussion, I may write a separate standalone article with more details on these concepts in the future\n", + "- In the meantime, if you are interested in comparing RLHF and DPO, please see the section [2.2. RLHF vs Direct Preference Optimization (DPO)](https://magazine.sebastianraschka.com/i/142924793/rlhf-vs-direct-preference-optimization-dpo) in my article [Tips for LLM Pretraining and Evaluating Reward Models](https://magazine.sebastianraschka.com/p/tips-for-llm-pretraining-and-evaluating-rms)" + ] + }, + { + "cell_type": "markdown", + "id": "xqVAgsyQ6LuG", + "metadata": { + "id": "xqVAgsyQ6LuG", + "tags": [] + }, + "source": [ + " \n", + "# 2) Preparing a preference dataset for DPO" + ] + }, + { + "cell_type": "markdown", + "id": "60b2195d-8734-469b-a52e-5031ca7ea6b1", + "metadata": { + "id": "60b2195d-8734-469b-a52e-5031ca7ea6b1" + }, + "source": [ + "- Let's begin by loading and preparing the dataset, which may already answer a lot of the questions you might have before we revisit the DPO loss equation\n", + "- Here, we work with a dataset that contains more polite and less polite responses to instruction prompts (concrete examples are shown in the next section)\n", + "- The dataset was generated via the [create-preference-data-ollama.ipynb](create-preference-data-ollama.ipynb) notebook" + ] + }, + { + "cell_type": "markdown", + "id": "wHLB62Nj7haD", + "metadata": { + "id": "wHLB62Nj7haD" + }, + "source": [ + " \n", + "## 2.1) Loading a preference dataset" + ] + }, + { + "cell_type": "markdown", + "id": "13e09f99-1b18-4923-ba36-af46d8e3075f", + "metadata": { + "id": "13e09f99-1b18-4923-ba36-af46d8e3075f" + }, + "source": [ + "- The dataset is a json file with 1100 entries:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "5266e66c-5ec0-45e6-a654-148971f6aee7", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "5266e66c-5ec0-45e6-a654-148971f6aee7", + "outputId": "04e8ee70-3076-441d-d2bf-7641da3d0c1d" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of entries: 1100\n" + ] + } + ], + "source": [ + "import json\n", + "\n", + "\n", + "file_path = \"instruction-data-with-preference.json\"\n", + "\n", + "with open(file_path, \"r\", encoding=\"utf-8\") as file:\n", + " data = json.load(file)\n", + "\n", + "print(\"Number of entries:\", len(data))" + ] + }, + { + "cell_type": "markdown", + "id": "725d2b9a-d6d2-46e2-89f8-5ab87e040e3b", + "metadata": { + "id": "725d2b9a-d6d2-46e2-89f8-5ab87e040e3b" + }, + "source": [ + "- Let's take a look at two example entries:" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "5c11916f-9a26-4367-a16e-7b0c121a20a6", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "5c11916f-9a26-4367-a16e-7b0c121a20a6", + "outputId": "00a432cc-19b1-484f-80e2-e897ee5e4024" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{'instruction': 'Identify the correct spelling of the following word.',\n", + " 'input': 'Ocassion',\n", + " 'output': \"The correct spelling is 'Occasion.'\",\n", + " 'rejected': \"The correct spelling is obviously 'Occasion.'\",\n", + " 'chosen': \"The correct spelling is 'Occasion.'\"}\n" + ] + } + ], + "source": [ + "import pprint\n", + "\n", + "pprint.pp(data[50])" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "01ef804a-8c13-4a0b-9b2e-b65a4d0a870d", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "01ef804a-8c13-4a0b-9b2e-b65a4d0a870d", + "outputId": "078cd643-83fb-4b42-ecf9-3256e8c9d239" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{'instruction': \"What is an antonym of 'complicated'?\",\n", + " 'input': '',\n", + " 'output': \"An antonym of 'complicated' is 'simple'.\",\n", + " 'chosen': \"A suitable antonym for 'complicated' would be 'simple'.\",\n", + " 'rejected': \"An antonym of 'complicated' is 'simple'.\"}\n" + ] + } + ], + "source": [ + "pprint.pp(data[999])" + ] + }, + { + "cell_type": "markdown", + "id": "56db5697-a089-4b40-a1f3-e928e8018220", + "metadata": { + "id": "56db5697-a089-4b40-a1f3-e928e8018220" + }, + "source": [ + "\n", + "\n", + "```\n", + "# This is formatted as code\n", + "```\n", + "\n", + "- As we can see above, the dataset consists of 5 keys:\n", + " - The `'instruction'` and `'input'` that are used as LLM inputs\n", + " - The `'output'` contains the response the model was trained on via the instruction finetuning step in chapter 7\n", + " - the `'chosen'` and `'rejected'` entries are the entries we use for DPO; here `'chosen'` is the preferred response, and `'rejected'` is the dispreferred response\n", + "- The goal is to get the model to follow the style of the chosen over the rejected responses" + ] + }, + { + "cell_type": "markdown", + "id": "86257468-a6ab-4ba3-9c9f-2fdc2c0cc284", + "metadata": { + "id": "86257468-a6ab-4ba3-9c9f-2fdc2c0cc284" + }, + "source": [ + "- Below is a utility function that formats the model input by applying the Alpaca prompt style similar to chapter 7 ([../01_main-chapter-code/ch07.ipynb](../01_main-chapter-code/ch07.ipynb)):" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "4564d55c-1c5d-46a6-b5e8-46ab568ad627", + "metadata": { + "id": "4564d55c-1c5d-46a6-b5e8-46ab568ad627" + }, + "outputs": [], + "source": [ + "def format_input(entry):\n", + " instruction_text = (\n", + " f\"Below is an instruction that describes a task. \"\n", + " f\"Write a response that appropriately completes the request.\"\n", + " f\"\\n\\n### Instruction:\\n{entry['instruction']}\"\n", + " )\n", + "\n", + " input_text = f\"\\n\\n### Input:\\n{entry['input']}\" if entry[\"input\"] else \"\"\n", + "\n", + " return instruction_text + input_text" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "3f38b49f-63fd-48c5-bde8-a4717b7923ea", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "3f38b49f-63fd-48c5-bde8-a4717b7923ea", + "outputId": "9ad07c59-05b3-42ae-c5bc-68780aaf6780" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Identify the correct spelling of the following word.\n", + "\n", + "### Input:\n", + "Ocassion\n" + ] + } + ], + "source": [ + "model_input = format_input(data[50])\n", + "print(model_input)" + ] + }, + { + "cell_type": "markdown", + "id": "7dd9e4c9-88a3-463a-8c16-c60ed7e6b51e", + "metadata": { + "id": "7dd9e4c9-88a3-463a-8c16-c60ed7e6b51e" + }, + "source": [ + "- Similarly, we can format the chosen and rejected responses using the Alpaca prompt style:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "8ad5831a-e936-44e5-a5cf-02953fe7d848", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "8ad5831a-e936-44e5-a5cf-02953fe7d848", + "outputId": "2c0a0cbf-c13d-43cf-fcc1-a4585c21e66f" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "### Response:\n", + "The correct spelling is 'Occasion.'\n" + ] + } + ], + "source": [ + "desired_response = f\"### Response:\\n{data[50]['chosen']}\"\n", + "print(desired_response)" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "fc0991f6-fef7-48ab-8dee-fbd2863f784c", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "fc0991f6-fef7-48ab-8dee-fbd2863f784c", + "outputId": "cd85406c-3470-48f8-9792-63f91affd50a" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "### Response:\n", + "The correct spelling is obviously 'Occasion.'\n" + ] + } + ], + "source": [ + "possible_response = f\"### Response:\\n{data[50]['rejected']}\"\n", + "print(possible_response)" + ] + }, + { + "cell_type": "markdown", + "id": "6G3j2Q987t_g", + "metadata": { + "id": "6G3j2Q987t_g" + }, + "source": [ + " \n", + "## 2.2) Creating training, validation, and test splits" + ] + }, + { + "cell_type": "markdown", + "id": "53ce2b1e-32d7-414c-8e6b-01f21a2488c2", + "metadata": { + "id": "53ce2b1e-32d7-414c-8e6b-01f21a2488c2" + }, + "source": [ + "- Next, we divide the dataset into 3 subsets, 85% training data, 5% validation data, and 10% test data:" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "36c7b919-8531-4e33-aebf-aaf8e6dbcfbd", + "metadata": { + "id": "36c7b919-8531-4e33-aebf-aaf8e6dbcfbd" + }, + "outputs": [], + "source": [ + "train_portion = int(len(data) * 0.85) # 85% for training\n", + "test_portion = int(len(data) * 0.1) # 10% for testing\n", + "val_portion = len(data) - train_portion - test_portion # Remaining 5% for validation\n", + "\n", + "train_data = data[:train_portion]\n", + "test_data = data[train_portion:train_portion + test_portion]\n", + "val_data = data[train_portion + test_portion:]" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "831a6c1b-119b-4622-9862-87f1db36e066", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "831a6c1b-119b-4622-9862-87f1db36e066", + "outputId": "8e017483-1a75-4336-9540-ac6a69104e27" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Training set length: 935\n", + "Validation set length: 55\n", + "Test set length: 110\n" + ] + } + ], + "source": [ + "print(\"Training set length:\", len(train_data))\n", + "print(\"Validation set length:\", len(val_data))\n", + "print(\"Test set length:\", len(test_data))" + ] + }, + { + "cell_type": "markdown", + "id": "c07d09f7-66af-49ed-8b9e-484f46e6a68d", + "metadata": { + "id": "c07d09f7-66af-49ed-8b9e-484f46e6a68d" + }, + "source": [ + " \n", + "## 2.3) Developing a `PreferenceDataset` class and batch processing function" + ] + }, + { + "cell_type": "markdown", + "id": "86101174-00c8-485d-8273-d086d5311926", + "metadata": { + "id": "86101174-00c8-485d-8273-d086d5311926" + }, + "source": [ + "- In this section, we rewrite the `InstructionDataset` class from chapter 7 ([../01_main-chapter-code/ch07.ipynb](../01_main-chapter-code/ch07.ipynb)) for DPO\n", + "- This means that instead of focusing on single output sequences (responses), we modify the dataset class to return pairs of responses where one is preferred (\"chosen\") over the other (\"rejected\")\n", + "- Overall, the `PreferenceDataset` is almost identical to the `InstructionDataset` used in chapter 7:" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "db08ad74-6dd4-4e40-b1e5-bc5f037d3d27", + "metadata": { + "id": "db08ad74-6dd4-4e40-b1e5-bc5f037d3d27" + }, + "outputs": [], + "source": [ + "import torch\n", + "from torch.utils.data import Dataset\n", + "\n", + "\n", + "class PreferenceDataset(Dataset):\n", + " def __init__(self, data, tokenizer):\n", + " self.data = data\n", + "\n", + " # Pre-tokenize texts\n", + " self.encoded_texts = []\n", + " for entry in data:\n", + " prompt = format_input(entry)\n", + " rejected_response = entry[\"rejected\"]\n", + " chosen_response = entry[\"chosen\"]\n", + "\n", + " prompt_tokens = tokenizer.encode(prompt)\n", + " chosen_full_text = f\"{prompt}\\n\\n### Response:\\n{chosen_response}\"\n", + " rejected_full_text = f\"{prompt}\\n\\n### Response:\\n{rejected_response}\"\n", + " chosen_full_tokens = tokenizer.encode(chosen_full_text)\n", + " rejected_full_tokens = tokenizer.encode(rejected_full_text)\n", + "\n", + " self.encoded_texts.append({\n", + " \"prompt\": prompt_tokens,\n", + " \"chosen\": chosen_full_tokens,\n", + " \"rejected\": rejected_full_tokens,\n", + " })\n", + "\n", + " def __getitem__(self, index):\n", + " return self.encoded_texts[index]\n", + "\n", + " def __len__(self):\n", + " return len(self.data)\n" + ] + }, + { + "cell_type": "markdown", + "id": "2325d183-75b9-400a-80ac-0b8d2f526561", + "metadata": { + "id": "2325d183-75b9-400a-80ac-0b8d2f526561" + }, + "source": [ + "- Along with an updated `PreferenceDataset` class, we also need an updated batch collation function that we use to pad the sequences in each batch to an equal length so that we can assemble them in batches\n", + "- I added comments to the code below to illustrate the process; however, it might be easiest to understand how it works by looking at the example inputs and outputs further below:" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "8d3a43a6-7704-4bff-9bbc-a38632374f30", + "metadata": { + "id": "8d3a43a6-7704-4bff-9bbc-a38632374f30" + }, + "outputs": [], + "source": [ + "def custom_collate_fn(\n", + " batch,\n", + " pad_token_id=50256,\n", + " allowed_max_length=None,\n", + " mask_prompt_tokens=True,\n", + " device=\"cpu\"\n", + "):\n", + " # Initialize lists to hold batch data\n", + " batch_data = {\n", + " \"prompt\": [],\n", + " \"chosen\": [],\n", + " \"rejected\": [],\n", + " \"rejected_mask\": [],\n", + " \"chosen_mask\": []\n", + "\n", + " }\n", + "\n", + " # Determine the longest sequence to set a common padding length\n", + " max_length_common = 0\n", + " if batch:\n", + " for key in [\"chosen\", \"rejected\"]:\n", + " current_max = max(len(item[key])+1 for item in batch)\n", + " max_length_common = max(max_length_common, current_max)\n", + "\n", + " # Process each item in the batch\n", + " for item in batch:\n", + " prompt = torch.tensor(item[\"prompt\"])\n", + " batch_data[\"prompt\"].append(prompt)\n", + "\n", + " for key in [\"chosen\", \"rejected\"]:\n", + " # Adjust padding according to the common maximum length\n", + " sequence = item[key]\n", + " padded = sequence + [pad_token_id] * (max_length_common - len(sequence))\n", + " mask = torch.ones(len(padded)).bool()\n", + "\n", + " # Set mask for all padding tokens to False\n", + " mask[len(sequence):] = False\n", + "\n", + " # Set mask for all input tokens to False\n", + " # +2 sets the 2 newline (\"\\n\") tokens before \"### Response\" to False\n", + " if mask_prompt_tokens:\n", + " mask[:prompt.shape[0]+2] = False\n", + "\n", + " batch_data[key].append(torch.tensor(padded))\n", + " batch_data[f\"{key}_mask\"].append(mask)\n", + "\n", + " # Final processing\n", + " for key in [\"chosen\", \"rejected\", \"chosen_mask\", \"rejected_mask\"]:\n", + " # Stack all sequences into a tensor for the given key\n", + " tensor_stack = torch.stack(batch_data[key])\n", + "\n", + " # Optionally truncate to maximum sequence length\n", + " if allowed_max_length is not None:\n", + " tensor_stack = tensor_stack[:, :allowed_max_length]\n", + "\n", + " # Move to the specified device\n", + " batch_data[key] = tensor_stack.to(device)\n", + "\n", + " return batch_data" + ] + }, + { + "cell_type": "markdown", + "id": "76f3744b-9bb0-4f1e-b66b-cff35ad8fd9f", + "metadata": { + "id": "76f3744b-9bb0-4f1e-b66b-cff35ad8fd9f" + }, + "source": [ + "- Before we start using the custom collate function, let's make version of it with some of its function arguments prefilled:" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "d3cc137c-7ed7-4758-a518-cc4071b2817a", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "d3cc137c-7ed7-4758-a518-cc4071b2817a", + "outputId": "598e9def-9768-441a-f886-01f6ba6e250b" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Device: cuda\n" + ] + } + ], + "source": [ + "from functools import partial\n", + "\n", + "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", + "print(\"Device:\", device)\n", + "\n", + "customized_collate_fn = partial(\n", + " custom_collate_fn,\n", + " device=device, # Put the data directly on a GPU if available\n", + " mask_prompt_tokens=True, # This is optional\n", + " allowed_max_length=1024 # The supported context length of the model\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "5d29e996-e267-4348-bc1d-4ac6b725cf6a", + "metadata": { + "id": "5d29e996-e267-4348-bc1d-4ac6b725cf6a" + }, + "source": [ + "- Now, let's see the `customized_collate_fn` in action and apply it to some sample data from our preference dataset; for this, we take the first two entries:" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "1171057d-2a0f-48ff-bad6-4917a072f0f5", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "1171057d-2a0f-48ff-bad6-4917a072f0f5", + "outputId": "3db3eee8-db29-4ff6-8078-6577a05d953a" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "{'instruction': 'Evaluate the following phrase by transforming it into the '\n", + " 'spelling given.',\n", + " 'input': 'freind --> friend',\n", + " 'output': 'The spelling of the given phrase \"freind\" is incorrect, the '\n", + " 'correct spelling is \"friend\".',\n", + " 'rejected': 'The spelling of the given phrase \"freind\" is flat out wrong, get '\n", + " 'it together, the correct spelling is \"friend\".',\n", + " 'chosen': 'The spelling of the given phrase \"freind\" is incorrect, the '\n", + " 'correct spelling is \"friend\".'}\n", + "\n", + "{'instruction': 'Edit the following sentence for grammar.',\n", + " 'input': 'He go to the park every day.',\n", + " 'output': 'He goes to the park every day.',\n", + " 'rejected': 'He goes to the stupid park every single day.',\n", + " 'chosen': 'He goes to the park every day.'}\n" + ] + } + ], + "source": [ + "example_data = data[:2]\n", + "\n", + "for i in example_data:\n", + " print()\n", + " pprint.pp(i)" + ] + }, + { + "cell_type": "markdown", + "id": "8f1436cc-fbe5-4581-89d8-1992b5f04042", + "metadata": { + "id": "8f1436cc-fbe5-4581-89d8-1992b5f04042" + }, + "source": [ + "- Next, let's instantiate an `example_dataset` and use a PyTorch `DataLoader` to create an `example_dataloader` that mimics the data loader we will use for the model training later:" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "db327575-c34b-4fea-b3c7-e30569c9be78", + "metadata": { + "id": "db327575-c34b-4fea-b3c7-e30569c9be78" + }, + "outputs": [], + "source": [ + "import tiktoken\n", + "from torch.utils.data import DataLoader\n", + "\n", + "\n", + "tokenizer = tiktoken.get_encoding(\"gpt2\")\n", + "\n", + "example_dataset = PreferenceDataset(example_data, tokenizer)\n", + "\n", + "example_dataloader = DataLoader(\n", + " example_dataset,\n", + " batch_size=2,\n", + " collate_fn=customized_collate_fn,\n", + " shuffle=False\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "43a446b7-7037-4d9a-9f14-b4ee0f6f37af", + "metadata": { + "id": "43a446b7-7037-4d9a-9f14-b4ee0f6f37af" + }, + "source": [ + "- The dataset has the following keys:" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "87ed4cf9-d70a-4bc7-b676-67e76ed3ee10", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "87ed4cf9-d70a-4bc7-b676-67e76ed3ee10", + "outputId": "fa724d65-b0e1-4239-8090-9263135ad199" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "batch.keys: dict_keys(['prompt', 'chosen', 'rejected', 'rejected_mask', 'chosen_mask'])\n" + ] + } + ], + "source": [ + "for batch in example_dataloader:\n", + " break\n", + "\n", + "print(\"batch.keys:\", batch.keys())" + ] + }, + { + "cell_type": "markdown", + "id": "5bda3193-8c68-478c-98d8-0d9d880e7077", + "metadata": { + "id": "5bda3193-8c68-478c-98d8-0d9d880e7077" + }, + "source": [ + "- The prompts are a list of tensors, where each tensor contains the token IDs for a given example; since we selected a batch size of 2, we have two lists of token ID tensors here:" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "468995ce-2906-498f-ac99-0a3f80d13d12", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "468995ce-2906-498f-ac99-0a3f80d13d12", + "outputId": "7f3df961-fcb5-4e49-9b0c-c99447c67cc1" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "[tensor([21106, 318, 281, 12064, 326, 8477, 257, 4876, 13, 19430,\n", + " 257, 2882, 326, 20431, 32543, 262, 2581, 13, 198, 198,\n", + " 21017, 46486, 25, 198, 36, 2100, 4985, 262, 1708, 9546,\n", + " 416, 25449, 340, 656, 262, 24993, 1813, 13, 198, 198,\n", + " 21017, 23412, 25, 198, 19503, 521, 14610, 1545]),\n", + " tensor([21106, 318, 281, 12064, 326, 8477, 257, 4876, 13, 19430,\n", + " 257, 2882, 326, 20431, 32543, 262, 2581, 13, 198, 198,\n", + " 21017, 46486, 25, 198, 18378, 262, 1708, 6827, 329, 23491,\n", + " 13, 198, 198, 21017, 23412, 25, 198, 1544, 467, 284,\n", + " 262, 3952, 790, 1110, 13])]" + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "batch[\"prompt\"]" + ] + }, + { + "cell_type": "markdown", + "id": "89cadebe-2516-4ae0-a71f-a8a623f2e1da", + "metadata": { + "id": "89cadebe-2516-4ae0-a71f-a8a623f2e1da" + }, + "source": [ + "- We don't really need the responses for training; what we need to feed to the model during training are the `\"chosen\"` and `\"rejected\"` entries\n", + "- The `\"chosen\"` and `\"rejected\"` response entries are padded so that we can stack them as tensors; similar to the prompts, these response texts are encoded into token IDs:" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "e8f49c56-3989-4fe9-81ac-6bb3cce1a5b8", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "e8f49c56-3989-4fe9-81ac-6bb3cce1a5b8", + "outputId": "ccc0bd06-6e85-4ee9-893b-d985f26a835d" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "tensor([[21106, 318, 281, 12064, 326, 8477, 257, 4876, 13, 19430,\n", + " 257, 2882, 326, 20431, 32543, 262, 2581, 13, 198, 198,\n", + " 21017, 46486, 25, 198, 36, 2100, 4985, 262, 1708, 9546,\n", + " 416, 25449, 340, 656, 262, 24993, 1813, 13, 198, 198,\n", + " 21017, 23412, 25, 198, 19503, 521, 14610, 1545, 198, 198,\n", + " 21017, 18261, 25, 198, 464, 24993, 286, 262, 1813, 9546,\n", + " 366, 19503, 521, 1, 318, 11491, 11, 262, 3376, 24993,\n", + " 318, 366, 6726, 1911, 50256, 50256, 50256, 50256, 50256, 50256,\n", + " 50256],\n", + " [21106, 318, 281, 12064, 326, 8477, 257, 4876, 13, 19430,\n", + " 257, 2882, 326, 20431, 32543, 262, 2581, 13, 198, 198,\n", + " 21017, 46486, 25, 198, 18378, 262, 1708, 6827, 329, 23491,\n", + " 13, 198, 198, 21017, 23412, 25, 198, 1544, 467, 284,\n", + " 262, 3952, 790, 1110, 13, 198, 198, 21017, 18261, 25,\n", + " 198, 1544, 2925, 284, 262, 3952, 790, 1110, 13, 50256,\n", + " 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256,\n", + " 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256, 50256,\n", + " 50256]], device='cuda:0')" + ] + }, + "execution_count": 19, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "batch[\"chosen\"]" + ] + }, + { + "cell_type": "markdown", + "id": "35a4cd6d-b2ad-45a6-b00a-ba5b720be4ea", + "metadata": { + "id": "35a4cd6d-b2ad-45a6-b00a-ba5b720be4ea" + }, + "source": [ + "- The token IDs above represent the model inputs, but in this format, they are hard to interpret for us humans\n", + "- So, let's implement a small utility function to convert them back into text so that we can inspect and interpret them more easily:" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "52ea54ba-32cb-4ecb-b38b-923f42fd4615", + "metadata": { + "id": "52ea54ba-32cb-4ecb-b38b-923f42fd4615" + }, + "outputs": [], + "source": [ + "def decode_tokens_from_batch(token_ids, tokenizer):\n", + " ids_in_python_list = token_ids.flatten().tolist()\n", + " return tokenizer.decode(ids_in_python_list)" + ] + }, + { + "cell_type": "markdown", + "id": "bc9dd0ce-1fd4-419c-833f-ea5a1f8d800d", + "metadata": { + "id": "bc9dd0ce-1fd4-419c-833f-ea5a1f8d800d" + }, + "source": [ + "- Let's apply the `decode_tokens_from_batch` utility function to the first prompt entry in the batch:" + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "id": "55ee481e-3e2c-4ff6-b614-8cb18eb16a41", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "55ee481e-3e2c-4ff6-b614-8cb18eb16a41", + "outputId": "17ddec15-a09d-45b5-b1e8-600cd59a9600" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Evaluate the following phrase by transforming it into the spelling given.\n", + "\n", + "### Input:\n", + "freind --> friend\n" + ] + } + ], + "source": [ + "text = decode_tokens_from_batch(\n", + " token_ids=batch[\"prompt\"][0], # [0] for the first entry in the batch\n", + " tokenizer=tokenizer,\n", + ")\n", + "print(text)" + ] + }, + { + "cell_type": "markdown", + "id": "637b95c4-d5c2-4492-9d19-a45b090eee7e", + "metadata": { + "id": "637b95c4-d5c2-4492-9d19-a45b090eee7e" + }, + "source": [ + "- As we can see above, the prompt was correctly formatted; let's now do the same for the `\"chosen\"` response:" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "id": "33a24f20-5ec3-4a89-b57a-52e997163d07", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "33a24f20-5ec3-4a89-b57a-52e997163d07", + "outputId": "e04366ee-3719-4b07-fcef-6e9dddc06310" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Evaluate the following phrase by transforming it into the spelling given.\n", + "\n", + "### Input:\n", + "freind --> friend\n", + "\n", + "### Response:\n", + "The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\".<|endoftext|><|endoftext|><|endoftext|><|endoftext|><|endoftext|><|endoftext|><|endoftext|>\n" + ] + } + ], + "source": [ + "text = decode_tokens_from_batch(\n", + " token_ids=batch[\"chosen\"][0],\n", + " tokenizer=tokenizer,\n", + ")\n", + "print(text)" + ] + }, + { + "cell_type": "markdown", + "id": "ac9fbdbd-1cff-401f-8e6c-cd98c134c0f2", + "metadata": { + "id": "ac9fbdbd-1cff-401f-8e6c-cd98c134c0f2" + }, + "source": [ + "- As we can see above, similar to instruction finetuning, the response that is passed to the model during training also contains the input prompt\n", + "- Also note that we included `<|endoftext|>` tokens as padding tokens, which are necessary so that we can extend the responses to a similar length to stack them as a batch\n", + "- Don't worry; the `<|endoftext|>` tokens will be ignored in the loss later so that they won't affect the training outcome\n", + "- Let's now also inspect the corresponding rejected response:" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "id": "db382be5-c727-4299-8597-c05424ba9308", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "db382be5-c727-4299-8597-c05424ba9308", + "outputId": "edbd8c4a-0528-4361-aeba-9b3c3bbde33b" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Evaluate the following phrase by transforming it into the spelling given.\n", + "\n", + "### Input:\n", + "freind --> friend\n", + "\n", + "### Response:\n", + "The spelling of the given phrase \"freind\" is flat out wrong, get it together, the correct spelling is \"friend\".<|endoftext|>\n" + ] + } + ], + "source": [ + "text = decode_tokens_from_batch(\n", + " token_ids=batch[\"rejected\"][0],\n", + " tokenizer=tokenizer,\n", + ")\n", + "print(text)" + ] + }, + { + "cell_type": "markdown", + "id": "715dc968-aa64-4388-b577-7c295831bdcf", + "metadata": { + "id": "715dc968-aa64-4388-b577-7c295831bdcf" + }, + "source": [ + "- In this case, as we can see above, the rejected response is a more impolite version of the chosen response (we don't want the model to generate impolite responses)\n", + "- Lastly, let's talk about the data masks: if you took a closer look at our custom collate function we implemented above, we created a `\"chosen_mask\"` and a `\"rejected_mask\"` for each dataset entry\n", + "- The masks have the same shape as the response entries, as shown below for the `\"chosen\"` entry:" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "id": "5c324eab-cf1d-4071-b3ba-797d8ec4d1da", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "5c324eab-cf1d-4071-b3ba-797d8ec4d1da", + "outputId": "742a5742-1bc0-4f74-9eb9-cbf81f936ecb" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "chosen inputs: torch.Size([81])\n", + "chosen mask: torch.Size([81])\n" + ] + } + ], + "source": [ + "print(\"chosen inputs:\", batch[\"chosen\"][0].shape)\n", + "print(\"chosen mask: \", batch[\"chosen_mask\"][0].shape)" + ] + }, + { + "cell_type": "markdown", + "id": "880e95f7-cfc3-4f5f-be5e-c279fba5f674", + "metadata": { + "id": "880e95f7-cfc3-4f5f-be5e-c279fba5f674" + }, + "source": [ + "- The contents of these masks are boolean (`True` and `False`) values:" + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "id": "da75b550-5da4-4292-9a7e-a05b842bdcb7", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "da75b550-5da4-4292-9a7e-a05b842bdcb7", + "outputId": "e5f012c3-33ba-4e6b-aa55-3e331865218f" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "tensor([False, False, False, False, False, False, False, False, False, False,\n", + " False, False, False, False, False, False, False, False, False, False,\n", + " False, False, False, False, False, False, False, False, False, False,\n", + " False, False, False, False, False, False, False, False, False, False,\n", + " False, False, False, False, False, False, False, False, False, False,\n", + " True, True, True, True, True, True, True, True, True, True,\n", + " True, True, True, True, True, True, True, True, True, True,\n", + " True, True, True, True, False, False, False, False, False, False,\n", + " False], device='cuda:0')" + ] + }, + "execution_count": 25, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "batch[\"chosen_mask\"][0]" + ] + }, + { + "cell_type": "markdown", + "id": "0e67b862-4430-4c99-9157-90955dde29b6", + "metadata": { + "id": "0e67b862-4430-4c99-9157-90955dde29b6" + }, + "source": [ + "- The `True` values denote token IDs that correspond to the actual response\n", + "- the `False` tokens correspond to token IDs that correspond to either prompt tokens (if we set `mask_prompt_tokens=True` in the `customized_collate_fn` function, which we previously did) or padding tokens\n", + "- Hence, we can use the mask as a selection mask to select only the token IDs that correspond to the response, that is, stripping all prompt and padding tokens, as we can see below:" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "id": "1114c6fe-524b-401c-b9fe-02260e6f0541", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "1114c6fe-524b-401c-b9fe-02260e6f0541", + "outputId": "6d99af1d-940a-4012-c5d9-21d463a66e40" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "### Response:\n", + "The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\".\n" + ] + } + ], + "source": [ + "text = decode_tokens_from_batch(\n", + " token_ids=batch[\"chosen\"][0][batch[\"chosen_mask\"][0]],\n", + " tokenizer=tokenizer,\n", + ")\n", + "print(text)" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "id": "a89f83a4-d16e-40d2-ba43-bd410affd967", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "a89f83a4-d16e-40d2-ba43-bd410affd967", + "outputId": "1d439c7e-c079-4594-d02a-fa83a3cb275d" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "### Response:\n", + "The spelling of the given phrase \"freind\" is flat out wrong, get it together, the correct spelling is \"friend\".\n" + ] + } + ], + "source": [ + "text = decode_tokens_from_batch(\n", + " token_ids=batch[\"rejected\"][0][batch[\"rejected_mask\"][0]],\n", + " tokenizer=tokenizer,\n", + ")\n", + "print(text)" + ] + }, + { + "cell_type": "markdown", + "id": "e525287f-137c-4d71-94ae-cfd6db7b057c", + "metadata": { + "id": "e525287f-137c-4d71-94ae-cfd6db7b057c" + }, + "source": [ + "- We will make use of this mask to ignore prompt and padding tokens when computing the DPO loss later" + ] + }, + { + "cell_type": "markdown", + "id": "jbafhM_R8z5q", + "metadata": { + "id": "jbafhM_R8z5q" + }, + "source": [ + " \n", + "## 2.4) Creating training, validation, and test set data loaders" + ] + }, + { + "cell_type": "markdown", + "id": "b3c29eb8-d1b9-4abe-a155-52b3270d759a", + "metadata": { + "id": "b3c29eb8-d1b9-4abe-a155-52b3270d759a" + }, + "source": [ + "- Above, we worked with a small example subsets from the preference dataset for illustration purposes\n", + "- Let's now create the actual training, validation, and test set data loaders\n", + "- This process is identical to creating the data loaders in the pretraining and instruction finetuning chapters and thus should be self-explanatory" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "id": "5c0068bf-bda0-4d9e-9f79-2fc4b94cbd1c", + "metadata": { + "id": "5c0068bf-bda0-4d9e-9f79-2fc4b94cbd1c" + }, + "outputs": [], + "source": [ + "from torch.utils.data import DataLoader\n", + "\n", + "\n", + "num_workers = 0\n", + "batch_size = 8\n", + "\n", + "torch.manual_seed(123)\n", + "\n", + "train_dataset = PreferenceDataset(train_data, tokenizer)\n", + "train_loader = DataLoader(\n", + " train_dataset,\n", + " batch_size=batch_size,\n", + " collate_fn=customized_collate_fn,\n", + " shuffle=True,\n", + " drop_last=True,\n", + " num_workers=num_workers\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "id": "2f4a257b-6835-4194-abe2-5831d6a44885", + "metadata": { + "id": "2f4a257b-6835-4194-abe2-5831d6a44885" + }, + "outputs": [], + "source": [ + "val_dataset = PreferenceDataset(val_data, tokenizer)\n", + "val_loader = DataLoader(\n", + " val_dataset,\n", + " batch_size=batch_size,\n", + " collate_fn=customized_collate_fn,\n", + " shuffle=False,\n", + " drop_last=False,\n", + " num_workers=num_workers\n", + ")\n", + "\n", + "test_dataset = PreferenceDataset(test_data, tokenizer)\n", + "test_loader = DataLoader(\n", + " test_dataset,\n", + " batch_size=batch_size,\n", + " collate_fn=customized_collate_fn,\n", + " shuffle=False,\n", + " drop_last=False,\n", + " num_workers=num_workers\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "1fe1ba19-a6d5-4a77-8283-7a17d7ec06e2", + "metadata": { + "id": "1fe1ba19-a6d5-4a77-8283-7a17d7ec06e2" + }, + "source": [ + "- Let's iterate through the data loader and take a look at the dataset shapes:" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "id": "80d61f15-facb-4eb8-a9be-6427887d24b2", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "80d61f15-facb-4eb8-a9be-6427887d24b2", + "outputId": "dacd3bdf-f069-4b36-da2c-d6c1c6cc5405" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Train loader:\n", + "torch.Size([8, 77]) torch.Size([8, 77])\n", + "torch.Size([8, 81]) torch.Size([8, 81])\n", + "torch.Size([8, 94]) torch.Size([8, 94])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 99]) torch.Size([8, 99])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 88]) torch.Size([8, 88])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 79]) torch.Size([8, 79])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 97]) torch.Size([8, 97])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 89]) torch.Size([8, 89])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n", + "torch.Size([8, 84]) torch.Size([8, 84])\n", + "torch.Size([8, 79]) torch.Size([8, 79])\n", + "torch.Size([8, 101]) torch.Size([8, 101])\n", + "torch.Size([8, 87]) torch.Size([8, 87])\n", + "torch.Size([8, 73]) torch.Size([8, 73])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 73]) torch.Size([8, 73])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 91]) torch.Size([8, 91])\n", + "torch.Size([8, 78]) torch.Size([8, 78])\n", + "torch.Size([8, 78]) torch.Size([8, 78])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 84]) torch.Size([8, 84])\n", + "torch.Size([8, 92]) torch.Size([8, 92])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 73]) torch.Size([8, 73])\n", + "torch.Size([8, 73]) torch.Size([8, 73])\n", + "torch.Size([8, 78]) torch.Size([8, 78])\n", + "torch.Size([8, 66]) torch.Size([8, 66])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 100]) torch.Size([8, 100])\n", + "torch.Size([8, 77]) torch.Size([8, 77])\n", + "torch.Size([8, 92]) torch.Size([8, 92])\n", + "torch.Size([8, 93]) torch.Size([8, 93])\n", + "torch.Size([8, 115]) torch.Size([8, 115])\n", + "torch.Size([8, 81]) torch.Size([8, 81])\n", + "torch.Size([8, 95]) torch.Size([8, 95])\n", + "torch.Size([8, 81]) torch.Size([8, 81])\n", + "torch.Size([8, 94]) torch.Size([8, 94])\n", + "torch.Size([8, 70]) torch.Size([8, 70])\n", + "torch.Size([8, 89]) torch.Size([8, 89])\n", + "torch.Size([8, 90]) torch.Size([8, 90])\n", + "torch.Size([8, 70]) torch.Size([8, 70])\n", + "torch.Size([8, 85]) torch.Size([8, 85])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 72]) torch.Size([8, 72])\n", + "torch.Size([8, 84]) torch.Size([8, 84])\n", + "torch.Size([8, 84]) torch.Size([8, 84])\n", + "torch.Size([8, 65]) torch.Size([8, 65])\n", + "torch.Size([8, 63]) torch.Size([8, 63])\n", + "torch.Size([8, 74]) torch.Size([8, 74])\n", + "torch.Size([8, 79]) torch.Size([8, 79])\n", + "torch.Size([8, 93]) torch.Size([8, 93])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 99]) torch.Size([8, 99])\n", + "torch.Size([8, 81]) torch.Size([8, 81])\n", + "torch.Size([8, 77]) torch.Size([8, 77])\n", + "torch.Size([8, 74]) torch.Size([8, 74])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 73]) torch.Size([8, 73])\n", + "torch.Size([8, 87]) torch.Size([8, 87])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 75]) torch.Size([8, 75])\n", + "torch.Size([8, 81]) torch.Size([8, 81])\n", + "torch.Size([8, 86]) torch.Size([8, 86])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 63]) torch.Size([8, 63])\n", + "torch.Size([8, 82]) torch.Size([8, 82])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 68]) torch.Size([8, 68])\n", + "torch.Size([8, 97]) torch.Size([8, 97])\n", + "torch.Size([8, 72]) torch.Size([8, 72])\n", + "torch.Size([8, 85]) torch.Size([8, 85])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 85]) torch.Size([8, 85])\n", + "torch.Size([8, 87]) torch.Size([8, 87])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 74]) torch.Size([8, 74])\n", + "torch.Size([8, 92]) torch.Size([8, 92])\n", + "torch.Size([8, 85]) torch.Size([8, 85])\n", + "torch.Size([8, 72]) torch.Size([8, 72])\n", + "torch.Size([8, 93]) torch.Size([8, 93])\n", + "torch.Size([8, 82]) torch.Size([8, 82])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 93]) torch.Size([8, 93])\n", + "torch.Size([8, 80]) torch.Size([8, 80])\n", + "torch.Size([8, 87]) torch.Size([8, 87])\n", + "torch.Size([8, 69]) torch.Size([8, 69])\n", + "torch.Size([8, 90]) torch.Size([8, 90])\n", + "torch.Size([8, 99]) torch.Size([8, 99])\n", + "torch.Size([8, 104]) torch.Size([8, 104])\n", + "torch.Size([8, 101]) torch.Size([8, 101])\n", + "torch.Size([8, 98]) torch.Size([8, 98])\n", + "torch.Size([8, 79]) torch.Size([8, 79])\n", + "torch.Size([8, 71]) torch.Size([8, 71])\n", + "torch.Size([8, 76]) torch.Size([8, 76])\n", + "torch.Size([8, 79]) torch.Size([8, 79])\n", + "torch.Size([8, 79]) torch.Size([8, 79])\n", + "torch.Size([8, 67]) torch.Size([8, 67])\n", + "torch.Size([8, 84]) torch.Size([8, 84])\n", + "torch.Size([8, 78]) torch.Size([8, 78])\n", + "torch.Size([8, 85]) torch.Size([8, 85])\n", + "torch.Size([8, 70]) torch.Size([8, 70])\n" + ] + } + ], + "source": [ + "print(\"Train loader:\")\n", + "for batch in train_loader:\n", + " print(\n", + " batch[\"chosen\"].shape,\n", + " batch[\"rejected\"].shape,\n", + " )" + ] + }, + { + "cell_type": "markdown", + "id": "7ff958a6-5e61-49f5-9a97-360aa34e3758", + "metadata": { + "id": "7ff958a6-5e61-49f5-9a97-360aa34e3758" + }, + "source": [ + "- Each row shows the shape of the `\"chosen\"` and `\"rejected\"` entries in each batch\n", + "- Since we applied padding on a batch-by-batch basis, each row has a different shape\n", + "- This is for efficiency reasons because it would be inefficient to pad all samples to the longest sample in the whole dataset" + ] + }, + { + "cell_type": "markdown", + "id": "29cb0543-1142-4374-8825-3384e20c6ac0", + "metadata": { + "id": "29cb0543-1142-4374-8825-3384e20c6ac0" + }, + "source": [ + " \n", + "# 3) Loading a finetuned LLM for DPO alignment" + ] + }, + { + "cell_type": "markdown", + "id": "22b08881-b769-4b26-8153-5ec0e8573ed2", + "metadata": { + "id": "22b08881-b769-4b26-8153-5ec0e8573ed2" + }, + "source": [ + "- LLM alignment steps, such as RLHF or DPO, assume that we already have an instruction-finetuned model\n", + "- This section contains minimal code to load the model that was instruction finetuned and saved in chapter 7 (via [../01_main-chapter-code/ch07.ipynb](../01_main-chapter-code/ch07.ipynb))\n", + "- Make sure you run the chapter 7 code first to create the instruction-finetuned model before you proceed\n", + "- The code below will copy the instruction-finetuned model into the current directory:" + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "id": "b3c6d82b-63f7-459a-b901-7125ab225e56", + "metadata": { + "id": "b3c6d82b-63f7-459a-b901-7125ab225e56" + }, + "outputs": [], + "source": [ + "import os\n", + "from pathlib import Path\n", + "import shutil\n", + "\n", + "\n", + "finetuned_model_path = Path(\"gpt2-medium355M-sft.pth\")\n", + "if not finetuned_model_path.exists():\n", + "\n", + " # Try finding the model checkpoint locally:\n", + " relative_path = Path(\"..\") / \"01_main-chapter-code\" / finetuned_model_path\n", + " if relative_path.exists():\n", + " shutil.copy(relative_path, \".\")\n", + "\n", + " # If this notebook is run on Google Colab, get it from a Google Drive folder\n", + " elif \"COLAB_GPU\" in os.environ or \"COLAB_TPU_ADDR\" in os.environ:\n", + " from google.colab import drive\n", + " drive.mount(\"/content/drive\")\n", + " google_drive_path = \"/content/drive/My Drive/Books/LLMs-From-Scratch/ch07/colab/gpt2-medium355M-sft.pth\" # Readers need to adjust this path\n", + " shutil.copy(google_drive_path, \".\")\n", + "\n", + " else:\n", + " print(\n", + " f\"Could not find '{finetuned_model_path}'.\\n\"\n", + " \"Run the `ch07.ipynb` notebook to finetune and save the finetuned model.\"\n", + " )" + ] + }, + { + "cell_type": "markdown", + "id": "71c8585e-4569-4033-84a7-3903d0e8aaf8", + "metadata": { + "id": "71c8585e-4569-4033-84a7-3903d0e8aaf8" + }, + "source": [ + "- Next, we reuse the basic configuration from previous chapters to load the model weights:" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "id": "a8333fee-e7fe-4f8c-9411-8c1db6252d98", + "metadata": { + "id": "a8333fee-e7fe-4f8c-9411-8c1db6252d98" + }, + "outputs": [], + "source": [ + "from previous_chapters import GPTModel\n", + "\n", + "\n", + "BASE_CONFIG = {\n", + " \"vocab_size\": 50257, # Vocabulary size\n", + " \"context_length\": 1024, # Context length\n", + " \"drop_rate\": 0.0, # Dropout rate\n", + " \"qkv_bias\": True # Query-key-value bias\n", + "}\n", + "\n", + "model_configs = {\n", + " \"gpt2-small (124M)\": {\"emb_dim\": 768, \"n_layers\": 12, \"n_heads\": 12},\n", + " \"gpt2-medium (355M)\": {\"emb_dim\": 1024, \"n_layers\": 24, \"n_heads\": 16},\n", + " \"gpt2-large (774M)\": {\"emb_dim\": 1280, \"n_layers\": 36, \"n_heads\": 20},\n", + " \"gpt2-xl (1558M)\": {\"emb_dim\": 1600, \"n_layers\": 48, \"n_heads\": 25},\n", + "}\n", + "\n", + "CHOOSE_MODEL = \"gpt2-medium (355M)\"\n", + "\n", + "BASE_CONFIG.update(model_configs[CHOOSE_MODEL])\n", + "\n", + "model = GPTModel(BASE_CONFIG)" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "id": "c2821403-605c-4071-a4ff-e23f4c9a11fd", + "metadata": { + "id": "c2821403-605c-4071-a4ff-e23f4c9a11fd" + }, + "outputs": [], + "source": [ + "model.load_state_dict(\n", + " torch.load(\n", + " \"gpt2-medium355M-sft.pth\",\n", + " map_location=torch.device(\"cpu\"),\n", + " weights_only=True\n", + " )\n", + ")\n", + "model.eval();" + ] + }, + { + "cell_type": "markdown", + "id": "61863bec-bd42-4194-b994-645bfe2df8be", + "metadata": { + "id": "61863bec-bd42-4194-b994-645bfe2df8be" + }, + "source": [ + "- Before training the loaded model with DPO, let's make sure that the finetuned model was saved and loaded correctly by trying it out on some sample data:" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "id": "4357aec5-0db2-4d73-b37b-539cd8fa80a3", + "metadata": { + "id": "4357aec5-0db2-4d73-b37b-539cd8fa80a3" + }, + "outputs": [], + "source": [ + "prompt = \"\"\"Below is an instruction that describes a task. Write a response\n", + "that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Convert the active sentence to passive: 'The chef cooks the meal every day.'\n", + "\"\"\"" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "id": "541e7988-38d3-47f6-bd52-9da6564479fa", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "541e7988-38d3-47f6-bd52-9da6564479fa", + "outputId": "278f7ddf-37c2-4c3a-d069-c510ef6f8d7a" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response\n", + "that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Convert the active sentence to passive: 'The chef cooks the meal every day.'\n", + "\n", + "### Response:\n", + "The meal is cooked every day by the chef.\n" + ] + } + ], + "source": [ + "from previous_chapters import (\n", + " generate,\n", + " text_to_token_ids,\n", + " token_ids_to_text\n", + ")\n", + "\n", + "torch.manual_seed(123)\n", + "\n", + "token_ids = generate(\n", + " model=model,\n", + " idx=text_to_token_ids(prompt, tokenizer),\n", + " max_new_tokens=35,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + ")\n", + "\n", + "response = token_ids_to_text(token_ids, tokenizer)\n", + "print(response)" + ] + }, + { + "cell_type": "markdown", + "id": "be87ed19-fded-4e56-8585-6c7c0367b354", + "metadata": { + "id": "be87ed19-fded-4e56-8585-6c7c0367b354" + }, + "source": [ + "- As we can see above, the model gives a reasonable and correct response\n", + "- As explained in chapter 7, in practice, we would clean up the response to only return the response text with the prompt and prompt style removed (similar to what you are familiar with from ChatGPT, for example):" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "id": "0c30c4e2-af84-4ab4-95d0-9641e32c1e7f", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "0c30c4e2-af84-4ab4-95d0-9641e32c1e7f", + "outputId": "70192bbe-fdf6-43eb-c673-f573f8c70156" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "The meal is cooked every day by the chef.\n" + ] + } + ], + "source": [ + "def extract_response(response_text, input_text):\n", + " return response_text[len(input_text):].replace(\"### Response:\", \"\").strip()\n", + "\n", + "response = extract_response(response, prompt)\n", + "print(response)" + ] + }, + { + "cell_type": "markdown", + "id": "80442cb9-83b1-46b8-bad0-7d44297ca52d", + "metadata": { + "id": "80442cb9-83b1-46b8-bad0-7d44297ca52d" + }, + "source": [ + "- Now, we are almost ready to get to the DPO part\n", + "- As mentioned at the beginning of this notebook, DPO works with two LLMs: a policy model (the LLM that we want to optimize) and a reference model (the original model that we keep unchanged)\n", + "- Below, we rename the `model` as `policy_model` and instantiate a second instance of the model we refer to as the `reference_model`" + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "id": "5d88cc3a-312e-4b29-bc6d-de8354c1eb9f", + "metadata": { + "id": "5d88cc3a-312e-4b29-bc6d-de8354c1eb9f" + }, + "outputs": [], + "source": [ + "policy_model = model\n", + "\n", + "reference_model = GPTModel(BASE_CONFIG)\n", + "reference_model.load_state_dict(\n", + " torch.load(\n", + " \"gpt2-medium355M-sft.pth\",\n", + " map_location=torch.device(\"cpu\"),\n", + " weights_only=True\n", + " )\n", + ")\n", + "reference_model.eval()\n", + "\n", + "policy_model.to(device)\n", + "reference_model.to(device);" + ] + }, + { + "cell_type": "markdown", + "id": "9c6c1469-0038-4914-8aa5-15b1f81877cc", + "metadata": { + "id": "9c6c1469-0038-4914-8aa5-15b1f81877cc" + }, + "source": [ + " \n", + "# 4) Coding the DPO Loss Function" + ] + }, + { + "cell_type": "markdown", + "id": "75dbe60c-e4ce-413e-beec-22eff0237d11", + "metadata": { + "id": "75dbe60c-e4ce-413e-beec-22eff0237d11" + }, + "source": [ + "- After we took care of the model loading and dataset preparation in the previous sections, we can now get to the fun part and code the DPO loss\n", + "- Note that the DPO loss code below is based on the method proposed in the [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://arxiv.org/abs/2305.18290) paper\n", + "- For reference, the core DPO equation is shown again below:\n", + "\n", + "\n", + "\n", + "- In the equation above,\n", + " - \"expected value\" $\\mathbb{E}$ is statistics jargon and stands for the average or mean value of the random variable (the expression inside the brackets)\n", + " - The $\\pi_{\\theta}$ variable is the so-called policy (a term borrowed from reinforcement learning) and represents the LLM we want to optimize; $\\pi_{ref}$ is a reference LLM, which is typically the original LLM before optimization (at the beginning of the training, $\\pi_{\\theta}$ and $\\pi_{ref}$ are typically the same)\n", + " - $\\beta$ is a hyperparameter to control the divergence between the $\\pi_{\\theta}$ and the reference model; increasing $\\beta$ increases the impact of the difference between\n", + "$\\pi_{\\theta}$ and $\\pi_{ref}$ in terms of their log probabilities on the overall loss function, thereby increasing the divergence between the two models\n", + "- In code, we can implement the DPO loss as follows:" + ] + }, + { + "cell_type": "code", + "execution_count": 38, + "id": "38CsrrwJIZiV", + "metadata": { + "id": "38CsrrwJIZiV" + }, + "outputs": [], + "source": [ + "import torch.nn.functional as F\n", + "\n", + "def compute_dpo_loss(\n", + " model_chosen_logprobs,\n", + " model_rejected_logprobs,\n", + " reference_chosen_logprobs,\n", + " reference_rejected_logprobs,\n", + " beta=0.1,\n", + " ):\n", + " \"\"\"Compute the DPO loss for a batch of policy and reference model log probabilities.\n", + "\n", + " Args:\n", + " policy_chosen_logprobs: Log probabilities of the policy model for the chosen responses. Shape: (batch_size,)\n", + " policy_rejected_logprobs: Log probabilities of the policy model for the rejected responses. Shape: (batch_size,)\n", + " reference_chosen_logprobs: Log probabilities of the reference model for the chosen responses. Shape: (batch_size,)\n", + " reference_rejected_logprobs: Log probabilities of the reference model for the rejected responses. Shape: (batch_size,)\n", + " beta: Temperature parameter for the DPO loss; typically something in the range of 0.1 to 0.5. We ignore the reference model as beta -> 0.\n", + " label_smoothing: conservativeness for DPO loss.\n", + "\n", + " Returns:\n", + " A tuple of three tensors: (loss, chosen_rewards, rejected_rewards).\n", + " \"\"\"\n", + "\n", + " model_logratios = model_chosen_logprobs - model_rejected_logprobs\n", + " reference_logratios = reference_chosen_logprobs - reference_rejected_logprobs\n", + " logits = model_logratios - reference_logratios\n", + "\n", + " # DPO (Eq. 7 of https://arxiv.org/pdf/2305.18290.pdf)\n", + " losses = -F.logsigmoid(beta * logits)\n", + "\n", + " # Optional values to track progress during training\n", + " chosen_rewards = (model_chosen_logprobs - reference_chosen_logprobs).detach()\n", + " rejected_rewards = (model_rejected_logprobs - reference_rejected_logprobs).detach()\n", + "\n", + " # .mean() to average over the samples in the batch\n", + " return losses.mean(), chosen_rewards.mean(), rejected_rewards.mean()" + ] + }, + { + "cell_type": "markdown", + "id": "693be65b-38fc-4d18-bf53-a260a15436e1", + "metadata": { + "id": "693be65b-38fc-4d18-bf53-a260a15436e1" + }, + "source": [ + "- If you are familiar with logarithms, note that we have the general relationship $\\log\\left(\\frac{a}{b}\\right) = \\log a - \\log b$, which we applied in the code above\n", + "- Keeping this in mind, let's go through some of the steps (we will calculate the `logprobs` using a separate function later)\n", + "- Let's start with the lines\n", + "\n", + " ```python\n", + " model_logratios = model_chosen_logprobs - model_rejected_logprobs\n", + " reference_logratios = reference_chosen_logprobs - reference_rejected_logprobs\n", + " ```\n", + "\n", + "- These lines above calculate the difference in log probabilities (logits) for the chosen and rejected samples for both the policy model and the reference model (this is due to $\\log\\left(\\frac{a}{b}\\right) = \\log a - \\log b$):\n", + "\n", + "$$\\log \\left( \\frac{\\pi_\\theta (y_w \\mid x)}{\\pi_\\theta (y_l \\mid x)} \\right) \\quad \\text{and} \\quad \\log \\left( \\frac{\\pi_{\\text{ref}}(y_w \\mid x)}{\\pi_{\\text{ref}}(y_l \\mid x)} \\right)$$" + ] + }, + { + "cell_type": "markdown", + "id": "5458d217-e0ad-40a5-925c-507a8fcf5795", + "metadata": { + "id": "5458d217-e0ad-40a5-925c-507a8fcf5795" + }, + "source": [ + "- Next, the code `logits = model_logratios - reference_logratios` computes the difference between the model's log ratios and the reference model's log ratios, i.e., \n", + "\n", + "$$\\beta \\log \\left( \\frac{\\pi_\\theta (y_w \\mid x)}{\\pi_{\\text{ref}} (y_w \\mid x)} \\right)\n", + "- \\beta \\log \\left( \\frac{\\pi_\\theta (y_l \\mid x)}{\\pi_{\\text{ref}} (y_l \\mid x)} \\right)$$\n" + ] + }, + { + "cell_type": "markdown", + "id": "f18e3e36-f5f1-407f-b662-4c20a0ac0354", + "metadata": { + "id": "f18e3e36-f5f1-407f-b662-4c20a0ac0354" + }, + "source": [ + "- Finally, `losses = -F.logsigmoid(beta * logits)` calculates the loss using the log-sigmoid function; in the original equation, the term inside the expectation is \n", + "\n", + "$$\\log \\sigma \\left( \\beta \\log \\left( \\frac{\\pi_\\theta (y_w \\mid x)}{\\pi_{\\text{ref}} (y_w \\mid x)} \\right)\n", + "- \\beta \\log \\left( \\frac{\\pi_\\theta (y_l \\mid x)}{\\pi_{\\text{ref}} (y_l \\mid x)} \\right) \\right)$$" + ] + }, + { + "cell_type": "markdown", + "id": "00a6f92d-7d64-41fe-bcaa-2bddd46027e1", + "metadata": { + "id": "00a6f92d-7d64-41fe-bcaa-2bddd46027e1" + }, + "source": [ + "- Above, we assumed that the log probabilities were already computed; let's now define a `compute_logprobs` function that we can use to compute these log probabilities that were passed into the `compute_dpo_loss` function above, that is, the values $\\pi_\\theta (y_w \\mid x)$, ${\\pi_\\theta (y_l \\mid x)}$, and so forth:" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "id": "71e6507b-d2e2-4469-86b9-f057b08b5df9", + "metadata": { + "id": "71e6507b-d2e2-4469-86b9-f057b08b5df9" + }, + "outputs": [], + "source": [ + "def compute_logprobs(logits, labels, selection_mask=None):\n", + " \"\"\"\n", + " Compute log probabilities.\n", + "\n", + " Args:\n", + " logits: Tensor of shape (batch_size, num_tokens, vocab_size)\n", + " labels: Tensor of shape (batch_size, num_tokens)\n", + " selection_mask: Tensor for shape (batch_size, num_tokens)\n", + "\n", + " Returns:\n", + " mean_log_prob: Mean log probability excluding padding tokens.\n", + " \"\"\"\n", + "\n", + " # Labels are the inputs shifted by one\n", + " labels = labels[:, 1:].clone()\n", + "\n", + " # Truncate logits to match the labels num_tokens\n", + " logits = logits[:, :-1, :]\n", + "\n", + " log_probs = F.log_softmax(logits, dim=-1)\n", + "\n", + " # Gather the log probabilities for the actual labels\n", + " selected_log_probs = torch.gather(\n", + " input=log_probs,\n", + " dim=-1,\n", + " index=labels.unsqueeze(-1)\n", + " ).squeeze(-1)\n", + "\n", + " if selection_mask is not None:\n", + " mask = selection_mask[:, 1:].clone()\n", + "\n", + " # Apply the mask to filter out padding tokens\n", + " selected_log_probs = selected_log_probs * mask\n", + "\n", + " # Calculate the average log probability excluding padding tokens\n", + " # This averages over the tokens, so the shape is (batch_size, num_tokens)\n", + " avg_log_prob = selected_log_probs.sum(-1) / mask.sum(-1)\n", + "\n", + " return avg_log_prob\n", + "\n", + " else:\n", + " return selected_log_probs.mean(-1)" + ] + }, + { + "cell_type": "markdown", + "id": "cf6a71ac-3fcc-44a4-befc-1c56bbd378d7", + "metadata": { + "id": "cf6a71ac-3fcc-44a4-befc-1c56bbd378d7" + }, + "source": [ + "- Note that this function above might look a bit intimidating at first due to the `torch.gather` function, but it's pretty similar to what happens under the hood in PyTorch's `cross_entropy` function\n", + "- For example, consider the following example:" + ] + }, + { + "cell_type": "code", + "execution_count": 40, + "id": "59873470-464d-4be2-860f-cbb7ac2d80ba", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "59873470-464d-4be2-860f-cbb7ac2d80ba", + "outputId": "8f7b47d4-73fe-4605-c17d-ad6cfd909a9b" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tensor(1.4185) tensor(1.4185)\n" + ] + } + ], + "source": [ + "# Sample data\n", + "logits = torch.tensor(\n", + " [[2.0, 1.0, 0.1],\n", + " [0.5, 2.5, 0.3]]) # Shape: (2, 3)\n", + "targets = torch.tensor([0, 2]) # Shape: (2,)\n", + "\n", + "\n", + "# Manual loss using torch.gather\n", + "log_softmax_logits = F.log_softmax(logits, dim=1) # Shape: (2, 3)\n", + "selected_log_probs = torch.gather(\n", + " input=log_softmax_logits,\n", + " dim=1,\n", + " index=targets.unsqueeze(1), # Shape 2, 1\n", + ").squeeze(1) # Shape: (2,)\n", + "manual_loss = -selected_log_probs.mean() # Averaging over the batch\n", + "\n", + "\n", + "# PyTorch loss\n", + "cross_entropy_loss = F.cross_entropy(logits, targets)\n", + "\n", + "print(manual_loss, cross_entropy_loss)" + ] + }, + { + "cell_type": "markdown", + "id": "f86d7add-f7ff-4a87-9193-7878c42bf0e7", + "metadata": { + "id": "f86d7add-f7ff-4a87-9193-7878c42bf0e7" + }, + "source": [ + "- So, above, we can see that the two implementations are equivalent, but let's narrow down a bit further to the `torch.gather` mechanics\n", + "- Consider the following two tensors:" + ] + }, + { + "cell_type": "code", + "execution_count": 41, + "id": "508db6ba-cc40-479f-a996-2250cf862388", + "metadata": { + "id": "508db6ba-cc40-479f-a996-2250cf862388" + }, + "outputs": [], + "source": [ + "t = torch.tensor(\n", + " [[1., 2.,],\n", + " [3., 4.]]\n", + ")\n", + "\n", + "m = torch.tensor(\n", + " [[1, 1],\n", + " [0, 1]]\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "821cbf45-8fbb-47b7-bae8-6c3271e36979", + "metadata": { + "id": "821cbf45-8fbb-47b7-bae8-6c3271e36979" + }, + "source": [ + "- Above, `t` is a tensor we want to select from, and `m` is a mask to specify how we want to select\n", + " - For instance, since `m` contains `[1, 1]` n the first row, it will select two times the value of `t` in index position `1`, which is the value 2.\n", + " - The second row of `m`, `[0, 1]`, selects index positions 0 and 1 in the second row or `t`, which are `3.` and `4.`" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "id": "4fdN5q1YPAbM", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "4fdN5q1YPAbM", + "outputId": "e935e8ad-1519-4c4b-dbff-65adae0a15a4" + }, + "outputs": [ + { + "data": { + "text/plain": [ + "tensor([[2., 2.],\n", + " [3., 4.]])" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "torch.gather(input=t, dim=-1, index=m)" + ] + }, + { + "cell_type": "markdown", + "id": "d10eeaf4-f24b-4e79-916a-abedf74fe4a3", + "metadata": { + "id": "d10eeaf4-f24b-4e79-916a-abedf74fe4a3" + }, + "source": [ + "- In other words, `torch.gather` is a selection function\n", + "- When we computed the loss earlier, we used it to retrieve the log probabilities corresponding to the correct token in the 50,256-token vocabulary\n", + "- The \"correct\" tokens are the tokens given in the response entry" + ] + }, + { + "cell_type": "markdown", + "id": "d5d10a43-ee5b-47ed-9d55-ddd96e66cf0b", + "metadata": { + "id": "d5d10a43-ee5b-47ed-9d55-ddd96e66cf0b" + }, + "source": [ + "- Regarding the `compute_logprobs` function above, we use `torch.gather` here because it gives us a bit more control than `cross_entropy`, but is, in essence, a similar idea\n", + "- The `selection_mask` we use there is to optionally ignore prompt and padding tokens\n", + "- We can then use the `compute_logprobs` function as follows to compute the inputs for the `compute_dpo_loss` loss function" + ] + }, + { + "cell_type": "code", + "execution_count": 43, + "id": "dfa7a4db-eba0-47d8-ad6d-7b5e7676e318", + "metadata": { + "id": "dfa7a4db-eba0-47d8-ad6d-7b5e7676e318" + }, + "outputs": [], + "source": [ + "def compute_dpo_loss_batch(batch, policy_model, reference_model, beta):\n", + " \"\"\"Compute the DPO loss on an input batch\"\"\"\n", + "\n", + " # where policy_model(batch[\"chosen\"]) are the logits\n", + " policy_chosen_log_probas = compute_logprobs(\n", + " logits=policy_model(batch[\"chosen\"]),\n", + " labels=batch[\"chosen\"],\n", + " selection_mask=batch[\"chosen_mask\"]\n", + " )\n", + " policy_rejected_log_probas = compute_logprobs(\n", + " logits=policy_model(batch[\"rejected\"]),\n", + " labels=batch[\"rejected\"],\n", + " selection_mask=batch[\"rejected_mask\"]\n", + " )\n", + " ref_chosen_log_probas = compute_logprobs(\n", + " logits=reference_model(batch[\"chosen\"]),\n", + " labels=batch[\"chosen\"],\n", + " selection_mask=batch[\"chosen_mask\"]\n", + " )\n", + " ref_rejected_log_probas = compute_logprobs(\n", + " logits=reference_model(batch[\"rejected\"]),\n", + " labels=batch[\"rejected\"],\n", + " selection_mask=batch[\"rejected_mask\"]\n", + " )\n", + " loss, chosen_rewards, rejected_rewards = compute_dpo_loss(\n", + " model_chosen_logprobs=policy_chosen_log_probas,\n", + " model_rejected_logprobs=policy_rejected_log_probas,\n", + " reference_chosen_logprobs=ref_chosen_log_probas,\n", + " reference_rejected_logprobs=ref_rejected_log_probas,\n", + " beta=beta\n", + " )\n", + " return loss, chosen_rewards, rejected_rewards" + ] + }, + { + "cell_type": "markdown", + "id": "b28caafb-f378-4332-a142-3e0f9ef67fbb", + "metadata": { + "id": "b28caafb-f378-4332-a142-3e0f9ef67fbb" + }, + "source": [ + "- The above function works for a single batch, for example:" + ] + }, + { + "cell_type": "code", + "execution_count": 44, + "id": "dd74fcc4-4280-41e9-9a22-838e85c84ee4", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "dd74fcc4-4280-41e9-9a22-838e85c84ee4", + "outputId": "65a70828-7dd2-4f72-ffec-45aeaf8afad0" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "(tensor(0.6931, device='cuda:0'), tensor(0., device='cuda:0'), tensor(0., device='cuda:0'))\n" + ] + } + ], + "source": [ + "with torch.no_grad():\n", + " loss = compute_dpo_loss_batch(batch, policy_model, reference_model, beta=0.1)\n", + "print(loss)" + ] + }, + { + "cell_type": "markdown", + "id": "b17429cd-2a00-41c8-9f16-38b1c9a5179f", + "metadata": { + "id": "b17429cd-2a00-41c8-9f16-38b1c9a5179f" + }, + "source": [ + "- Below, we extend this function to work for a specified `num_batches` in a data loader:" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "id": "682e9ad5-c5de-4d1b-9e93-3918bf5d5302", + "metadata": { + "id": "682e9ad5-c5de-4d1b-9e93-3918bf5d5302" + }, + "outputs": [], + "source": [ + "def compute_dpo_loss_loader(data_loader, policy_model, reference_model, beta, num_batches=None):\n", + " \"\"\"Apply compute_dpo_loss_batch to a whole data loader\"\"\"\n", + "\n", + " total_loss, total_chosen_rewards, total_rejected_rewards = 0., 0., 0.\n", + " if len(data_loader) == 0:\n", + " return float(\"nan\")\n", + "\n", + " elif num_batches is None:\n", + " num_batches = len(data_loader)\n", + " else:\n", + " # Reduce the number of batches to match the total number of batches in the data loader\n", + " # if num_batches exceeds the number of batches in the data loader\n", + " num_batches = min(num_batches, len(data_loader))\n", + " for i, batch in enumerate(data_loader):\n", + " if i < num_batches:\n", + " loss, chosen_rewards, rejected_rewards = compute_dpo_loss_batch(\n", + " batch=batch,\n", + " policy_model=policy_model,\n", + " reference_model=reference_model,\n", + " beta=beta\n", + " )\n", + " total_loss += loss.item()\n", + " total_chosen_rewards += chosen_rewards.item()\n", + " total_rejected_rewards += rejected_rewards.item()\n", + "\n", + " else:\n", + " break\n", + "\n", + " # calculate average\n", + " total_loss /= num_batches\n", + " total_chosen_rewards /= num_batches\n", + " total_rejected_rewards /= num_batches\n", + " return total_loss, total_chosen_rewards, total_rejected_rewards" + ] + }, + { + "cell_type": "markdown", + "id": "852e4c09-d285-44d5-be12-d29769950cb6", + "metadata": { + "id": "852e4c09-d285-44d5-be12-d29769950cb6" + }, + "source": [ + "- Why a specified `num_batches`? That's purely for efficiency reasons (because calculating the loss on the whole dataset each time would slow down the training significantly)" + ] + }, + { + "cell_type": "markdown", + "id": "2cca95b7-18fe-4076-9138-f70f21607b8c", + "metadata": { + "id": "2cca95b7-18fe-4076-9138-f70f21607b8c" + }, + "source": [ + "- Lastly, we define a convenience function for our training function later; this `evaluate_dpo_loss_loader` function computes the DPO loss and rewards for both the training and validation loader for logging purposes:" + ] + }, + { + "cell_type": "code", + "execution_count": 46, + "id": "c3d214ec-49ba-4bf0-ac80-f90fa0d832e9", + "metadata": { + "id": "c3d214ec-49ba-4bf0-ac80-f90fa0d832e9" + }, + "outputs": [], + "source": [ + "def evaluate_dpo_loss_loader(policy_model, reference_model, train_loader, val_loader, beta, eval_iter):\n", + " \"\"\"Compute the DPO loss for the training and validation dataset\"\"\"\n", + "\n", + " policy_model.eval()\n", + " with torch.no_grad():\n", + " train_loss, train_chosen_rewards, train_rejected_rewards = compute_dpo_loss_loader(\n", + " data_loader=train_loader,\n", + " policy_model=policy_model,\n", + " reference_model=reference_model,\n", + " beta=beta,\n", + " num_batches=eval_iter\n", + " )\n", + "\n", + " val_loss, val_chosen_rewards, val_rejected_rewards = compute_dpo_loss_loader(\n", + " data_loader=val_loader,\n", + " policy_model=policy_model,\n", + " reference_model=reference_model,\n", + " beta=beta,\n", + " num_batches=eval_iter\n", + " )\n", + "\n", + " res = {\n", + " \"train_loss\": train_loss,\n", + " \"train_chosen_reward\": train_chosen_rewards,\n", + " \"train_rejected_reward\": train_rejected_rewards,\n", + " \"val_loss\": val_loss,\n", + " \"val_chosen_reward\": val_chosen_rewards,\n", + " \"val_rejected_reward\": val_rejected_rewards\n", + " }\n", + "\n", + " policy_model.train()\n", + " return res" + ] + }, + { + "cell_type": "markdown", + "id": "6e95ed92-6743-4f13-8b91-0fbf2e540de1", + "metadata": { + "id": "6e95ed92-6743-4f13-8b91-0fbf2e540de1" + }, + "source": [ + "- In this section, we covered a lot of ground as a brief recap:\n", + " - The flow is: compute `logits` via the models $\\rightarrow$ `compute_logprobs` from logits $\\rightarrow$ compute `compute_dpo_loss` from log probabilities\n", + " - we have the `compute_dpo_loss_batch` function that facilitates the process above\n", + " - the `compute_dpo_loss_loader` utility function applies the `compute_dpo_loss_batch` function to a data loader\n", + " - the `evaluate_dpo_loss_loader` function applies the `compute_dpo_loss_batch` to both the training and validation set data loaders for logging purposes" + ] + }, + { + "cell_type": "markdown", + "id": "cb8a8f18-536e-4d83-a0d0-ac518a85f157", + "metadata": { + "id": "cb8a8f18-536e-4d83-a0d0-ac518a85f157" + }, + "source": [ + " \n", + "# 5) Training the model" + ] + }, + { + "cell_type": "markdown", + "id": "4b11d63d-3ddc-4070-9b2b-5ca0edb08d0c", + "metadata": { + "id": "4b11d63d-3ddc-4070-9b2b-5ca0edb08d0c" + }, + "source": [ + "- After setting up the DPO loss functions in the previous section, we can now finally train the model\n", + "- Note that this training function is the same one we used for pretraining and instruction finetuning, with minor differences:\n", + " - we swap the cross-entropy loss with our new DPO loss function\n", + " - we also track the rewards and reward margins, which are commonly used in RLHF and DPO contexts to track the training progress\n" + ] + }, + { + "cell_type": "code", + "execution_count": 47, + "id": "f90d9325-77b2-417f-88ff-0a5174889413", + "metadata": { + "id": "f90d9325-77b2-417f-88ff-0a5174889413" + }, + "outputs": [], + "source": [ + "from previous_chapters import generate_and_print_sample\n", + "\n", + "\n", + "def train_model_dpo_simple(\n", + " policy_model, reference_model, train_loader, val_loader,\n", + " optimizer, num_epochs, beta,\n", + " eval_freq, eval_iter, start_context, tokenizer\n", + "):\n", + "\n", + " # Initialize lists to track losses and tokens seen\n", + " tracking = {\n", + " \"train_losses\": [],\n", + " \"train_chosen_rewards\": [],\n", + " \"train_rejected_rewards\": [],\n", + " \"val_losses\": [],\n", + " \"val_chosen_rewards\": [],\n", + " \"val_rejected_rewards\": [],\n", + " \"tokens_seen\": []\n", + " }\n", + " tokens_seen, global_step = 0, -1\n", + "\n", + " # Main training loop\n", + " for epoch in range(num_epochs):\n", + " policy_model.train() # Set model to training mode\n", + "\n", + " for batch_idx, batch in enumerate(train_loader):\n", + "\n", + " optimizer.zero_grad() # Reset loss gradients from previous batch iteration\n", + "\n", + " loss, chosen_rewards, rejected_rewards = compute_dpo_loss_batch(\n", + " batch=batch,\n", + " policy_model=policy_model,\n", + " reference_model=reference_model,\n", + " beta=beta\n", + " )\n", + "\n", + " loss.backward() # Calculate loss gradients\n", + " optimizer.step() # Update model weights using loss gradients\n", + "\n", + " tokens_seen += batch[\"chosen\"].numel()\n", + " global_step += 1\n", + "\n", + " # Optional evaluation step\n", + " if global_step % eval_freq == 0:\n", + " res = evaluate_dpo_loss_loader(\n", + " policy_model=policy_model,\n", + " reference_model=reference_model,\n", + " train_loader=train_loader,\n", + " val_loader=val_loader,\n", + " beta=beta,\n", + " eval_iter=eval_iter\n", + " )\n", + " tracking[\"train_losses\"].append(res[\"train_loss\"])\n", + " tracking[\"train_chosen_rewards\"].append(res[\"train_chosen_reward\"])\n", + " tracking[\"train_rejected_rewards\"].append(res[\"train_rejected_reward\"])\n", + " tracking[\"val_losses\"].append(res[\"val_loss\"])\n", + " tracking[\"val_chosen_rewards\"].append(res[\"val_chosen_reward\"])\n", + " tracking[\"val_rejected_rewards\"].append(res[\"val_rejected_reward\"])\n", + " tracking[\"tokens_seen\"].append(tokens_seen)\n", + " train_reward_margin = res[\"train_chosen_reward\"] - res[\"train_rejected_reward\"]\n", + " val_reward_margin = res[\"val_chosen_reward\"] - res[\"val_rejected_reward\"]\n", + "\n", + " print(\n", + " f\"Ep {epoch+1} (Step {global_step:06d}): \"\n", + " f\"Train loss {res['train_loss']:.3f}, Val loss {res['val_loss']:.3f}, \"\n", + " f\"Train reward margins {train_reward_margin:.3f}, \"\n", + " f\"Val reward margins {val_reward_margin:.3f}\"\n", + " )\n", + "\n", + " # Print a sample text after each epoch\n", + " generate_and_print_sample(\n", + " model=model,\n", + " tokenizer=tokenizer,\n", + " device=loss.device,\n", + " start_context=start_context\n", + " )\n", + "\n", + " return tracking" + ] + }, + { + "cell_type": "markdown", + "id": "820d4904-f819-4d62-bfb4-85cf28863683", + "metadata": { + "id": "820d4904-f819-4d62-bfb4-85cf28863683" + }, + "source": [ + "- Before we start the training, let's print the initial losses and rewards:" + ] + }, + { + "cell_type": "code", + "execution_count": 48, + "id": "d53210c5-6d9c-46b0-af22-ee875c2806c5", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "d53210c5-6d9c-46b0-af22-ee875c2806c5", + "outputId": "8b1d2b39-16c5-4b99-e920-5b33d3c0f34d" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Training loss: 0.6931471824645996\n", + "Validation loss: 0.6931471824645996\n", + "Train reward margin: 0.0\n", + "Val reward margin: 0.0\n" + ] + } + ], + "source": [ + "torch.manual_seed(123) # For reproducibility due to the shuffling in the data loader\n", + "\n", + "res = evaluate_dpo_loss_loader(\n", + " policy_model=policy_model,\n", + " reference_model=reference_model,\n", + " train_loader=train_loader,\n", + " val_loader=val_loader,\n", + " beta=0.1,\n", + " eval_iter=5\n", + ")\n", + "\n", + "print(\"Training loss:\", res[\"train_loss\"])\n", + "print(\"Validation loss:\", res[\"val_loss\"])\n", + "\n", + "print(\"Train reward margin:\", res[\"train_chosen_reward\"] - res[\"train_rejected_reward\"])\n", + "print(\"Val reward margin:\", res[\"val_chosen_reward\"] - res[\"val_rejected_reward\"])" + ] + }, + { + "cell_type": "markdown", + "id": "4a006e91-df94-43ca-8025-1ba791e37bc4", + "metadata": { + "id": "4a006e91-df94-43ca-8025-1ba791e37bc4" + }, + "source": [ + "- Also, let's take a look at some of the initial model responses (the first 3 examples in the validation set):" + ] + }, + { + "cell_type": "code", + "execution_count": 49, + "id": "q4Ro9DrBa7zH", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "q4Ro9DrBa7zH", + "outputId": "b974d4bd-b92a-4a2a-bb7a-5a2a0d1eca11" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Convert the active sentence to passive: 'The chef cooks the meal every day.'\n", + "\n", + "Correct response:\n", + ">> The meal is cooked by the chef every day.\n", + "\n", + "Model response:\n", + ">> The meal is cooked every day by the chef.\n", + "\n", + "-------------------------------------\n", + "\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Classify an input string as either a noun or a verb.\n", + "\n", + "### Input:\n", + "Dance\n", + "\n", + "Correct response:\n", + ">> 'Dance' can be classified as a verb.\n", + "\n", + "Model response:\n", + ">> \"Dance\" can be classified as a verb.\n", + "\n", + "-------------------------------------\n", + "\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Rewrite the sentence using a metaphor.\n", + "\n", + "### Input:\n", + "The book is very interesting.\n", + "\n", + "Correct response:\n", + ">> The book is a page-turner.\n", + "\n", + "Model response:\n", + ">> The book is a treat.\n", + "\n", + "-------------------------------------\n", + "\n" + ] + } + ], + "source": [ + "torch.manual_seed(123)\n", + "\n", + "\n", + "for entry in val_data[:3]:\n", + "\n", + " input_text = format_input(entry)\n", + "\n", + " token_ids = generate(\n", + " model=model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + " response_text = (\n", + " generated_text[len(input_text):]\n", + " .replace(\"### Response:\", \"\")\n", + " .strip()\n", + ")\n", + "\n", + " print(input_text)\n", + " print(f\"\\nCorrect response:\\n>> {entry['output']}\")\n", + " print(f\"\\nModel response:\\n>> {response_text.strip()}\")\n", + " print(\"\\n-------------------------------------\\n\")" + ] + }, + { + "cell_type": "markdown", + "id": "ac2386ae-5c4c-448e-bfbf-4ec0604b171e", + "metadata": { + "id": "ac2386ae-5c4c-448e-bfbf-4ec0604b171e" + }, + "source": [ + "- Above, we see the original model responses\n", + "- Note that the goal of DPO is to induce slight style changes; this means we want the model to generate similar but slightly more polite responses\n", + "- Before we execute the following code cell that starts the training, here are a few notes about some of the settings:\n", + " - we are only passing the parameters of the policy model into the `AdamW` optimizer; that's the model we want to optimize (we don't want to modify the reference model)\n", + " - we only train for 1 epoch; that's because DPO is very prone to collapse (the loss might improve, but the model will start generating nonsensical texts)\n", + " - in DPO, it's best to use a very small learning rate\n", + " - the beta value can be increased from 0.1 to 0.5 to reduce the effect of DPO (we use 0.1 here to make the results more noticeable)\n", + " - The training takes about 2 minutes on an A100 GPU, but it can also be trained in 4 minutes on a smaller L4 GPU; training on a M3 MacBook Air takes about 30 minutes" + ] + }, + { + "cell_type": "code", + "execution_count": 50, + "id": "54b739be-871e-4c97-bf14-ffd2c58e1311", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "54b739be-871e-4c97-bf14-ffd2c58e1311", + "outputId": "d98b08b0-c325-411e-a1a4-05e7403f0345" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Ep 1 (Step 000000): Train loss 0.692, Val loss 0.693, Train reward margins 0.019, Val reward margins 0.009\n", + "Ep 1 (Step 000005): Train loss 0.690, Val loss 0.691, Train reward margins 0.070, Val reward margins 0.052\n", + "Ep 1 (Step 000010): Train loss 0.687, Val loss 0.688, Train reward margins 0.126, Val reward margins 0.108\n", + "Ep 1 (Step 000015): Train loss 0.676, Val loss 0.685, Train reward margins 0.362, Val reward margins 0.173\n", + "Ep 1 (Step 000020): Train loss 0.676, Val loss 0.680, Train reward margins 0.351, Val reward margins 0.264\n", + "Ep 1 (Step 000025): Train loss 0.666, Val loss 0.676, Train reward margins 0.564, Val reward margins 0.359\n", + "Ep 1 (Step 000030): Train loss 0.672, Val loss 0.672, Train reward margins 0.456, Val reward margins 0.441\n", + "Ep 1 (Step 000035): Train loss 0.663, Val loss 0.669, Train reward margins 0.658, Val reward margins 0.511\n", + "Ep 1 (Step 000040): Train loss 0.666, Val loss 0.666, Train reward margins 0.597, Val reward margins 0.574\n", + "Ep 1 (Step 000045): Train loss 0.648, Val loss 0.662, Train reward margins 0.982, Val reward margins 0.660\n", + "Ep 1 (Step 000050): Train loss 0.648, Val loss 0.659, Train reward margins 0.993, Val reward margins 0.734\n", + "Ep 1 (Step 000055): Train loss 0.647, Val loss 0.656, Train reward margins 1.014, Val reward margins 0.799\n", + "Ep 1 (Step 000060): Train loss 0.652, Val loss 0.653, Train reward margins 0.893, Val reward margins 0.870\n", + "Ep 1 (Step 000065): Train loss 0.631, Val loss 0.650, Train reward margins 1.361, Val reward margins 0.948\n", + "Ep 1 (Step 000070): Train loss 0.618, Val loss 0.646, Train reward margins 1.699, Val reward margins 1.038\n", + "Ep 1 (Step 000075): Train loss 0.617, Val loss 0.642, Train reward margins 1.733, Val reward margins 1.121\n", + "Ep 1 (Step 000080): Train loss 0.592, Val loss 0.639, Train reward margins 2.333, Val reward margins 1.194\n", + "Ep 1 (Step 000085): Train loss 0.610, Val loss 0.636, Train reward margins 1.907, Val reward margins 1.275\n", + "Ep 1 (Step 000090): Train loss 0.650, Val loss 0.633, Train reward margins 0.964, Val reward margins 1.353\n", + "Ep 1 (Step 000095): Train loss 0.607, Val loss 0.630, Train reward margins 1.962, Val reward margins 1.423\n", + "Ep 1 (Step 000100): Train loss 0.600, Val loss 0.627, Train reward margins 2.127, Val reward margins 1.500\n", + "Ep 1 (Step 000105): Train loss 0.590, Val loss 0.624, Train reward margins 2.458, Val reward margins 1.564\n", + "Ep 1 (Step 000110): Train loss 0.607, Val loss 0.622, Train reward margins 1.976, Val reward margins 1.621\n", + "Ep 1 (Step 000115): Train loss 0.621, Val loss 0.620, Train reward margins 1.605, Val reward margins 1.682\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request. ### Instruction: Rewrite the sentence using a metaphor. ### Input: The book is very interesting. ### Response: The book is a treat.<|endoftext|>The following is an instruction that describes a task. Write a response that appropriately completes the request. ### Input: The assignment was written by the student. ### Response\n", + "Training completed in 1.69 minutes.\n" + ] + } + ], + "source": [ + "import time\n", + "\n", + "start_time = time.time()\n", + "\n", + "torch.manual_seed(123)\n", + "\n", + "\n", + "optimizer = torch.optim.AdamW(policy_model.parameters(), lr=5e-6, weight_decay=0.01)\n", + "\n", + "num_epochs = 1\n", + "tracking = train_model_dpo_simple(\n", + " policy_model=policy_model,\n", + " reference_model=reference_model,\n", + " train_loader=train_loader,\n", + " val_loader=val_loader,\n", + " optimizer=optimizer,\n", + " num_epochs=num_epochs,\n", + " beta=0.1, # value between 0.1 and 0.5\n", + " eval_freq=5,\n", + " eval_iter=5,\n", + " start_context=format_input(val_data[2]),\n", + " tokenizer=tokenizer\n", + ")\n", + "\n", + "end_time = time.time()\n", + "execution_time_minutes = (end_time - start_time) / 60\n", + "print(f\"Training completed in {execution_time_minutes:.2f} minutes.\")" + ] + }, + { + "cell_type": "markdown", + "id": "eba8ea88-8771-4eb9-855d-2fe1ca2dc2fa", + "metadata": { + "id": "eba8ea88-8771-4eb9-855d-2fe1ca2dc2fa" + }, + "source": [ + "- As we can see based on the tracked results above, the loss improves\n", + "- Also, the reward margins, which is the difference between the rewards of the chosen and the rejected responses, improve, which is a good sign\n", + "- Let's take a more concrete look at these results in the next section" + ] + }, + { + "cell_type": "markdown", + "id": "11e23989-92bd-4ac2-a4bc-65d4c7ac334e", + "metadata": { + "id": "11e23989-92bd-4ac2-a4bc-65d4c7ac334e" + }, + "source": [ + " \n", + "# 6) Analyzing the results" + ] + }, + { + "cell_type": "markdown", + "id": "66d7d5fe-c617-45cb-8ea9-ddc7baa22654", + "metadata": { + "id": "66d7d5fe-c617-45cb-8ea9-ddc7baa22654" + }, + "source": [ + "- Let's begin analyzing the results by plotting the DPO loss:" + ] + }, + { + "cell_type": "code", + "execution_count": 51, + "id": "8ddcc66f-cd7c-4f46-96ea-af919ea1a199", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 307 + }, + "id": "8ddcc66f-cd7c-4f46-96ea-af919ea1a199", + "outputId": "c7164b26-8d32-41d1-8c6a-ab835d58d4c5" + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from previous_chapters import plot_losses\n", + "\n", + "\n", + "epochs_tensor = torch.linspace(0, num_epochs, len(tracking[\"train_losses\"]))\n", + "plot_losses(\n", + " epochs_seen=epochs_tensor,\n", + " tokens_seen=tracking[\"tokens_seen\"],\n", + " train_losses=tracking[\"train_losses\"],\n", + " val_losses=tracking[\"val_losses\"],\n", + " label=\"loss\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "7f8bc233-895f-46d5-8e01-202b991cd60c", + "metadata": { + "id": "7f8bc233-895f-46d5-8e01-202b991cd60c" + }, + "source": [ + "- As we can see above, the loss continues to improve, which is a good sign\n", + "- Based on the downward slope, one might be tempted to train the model a bit further (and readers are encouraged to try this), but not that DPO is prone to collapse, where the model may start generating nonsensical responses\n", + "- Next, let's take a look at the reward margins:" + ] + }, + { + "cell_type": "code", + "execution_count": 52, + "id": "dmbq6ruuf0Cl", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 307 + }, + "id": "dmbq6ruuf0Cl", + "outputId": "c2886c16-57da-41bd-c9f0-e936da9d9e4d" + }, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "train_reward_margins = [i-j for i,j in zip(tracking[\"train_chosen_rewards\"], tracking[\"train_rejected_rewards\"])]\n", + "val_reward_margins = [i-j for i,j in zip(tracking[\"val_chosen_rewards\"], tracking[\"val_rejected_rewards\"])]\n", + "\n", + "plot_losses(\n", + " epochs_seen=epochs_tensor,\n", + " tokens_seen=tracking[\"tokens_seen\"],\n", + " train_losses=train_reward_margins,\n", + " val_losses=val_reward_margins,\n", + " label=\"loss\"\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "69756011-acd6-404c-a5fc-7fe252cf20c8", + "metadata": { + "id": "69756011-acd6-404c-a5fc-7fe252cf20c8" + }, + "source": [ + "- As we can see, and as it's desired, the reward margins improve; this mirrors the loss curve and is a good sign\n", + "- Note that DPO losses and reward margins are valuable metrics to track during training; however, they don't tell the whole store\n", + "- Lastly, and most importantly, we have to conduct a qualitative check of the responses\n", + "- Here, we will look at the response (in addition, you could use an LLM to score the responses similar to chapter 7)" + ] + }, + { + "cell_type": "code", + "execution_count": 53, + "id": "5EfUXJGOali8", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "5EfUXJGOali8", + "outputId": "7ec7db47-d775-4646-f660-0d7f7e7c8503" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Convert the active sentence to passive: 'The chef cooks the meal every day.'\n", + "\n", + "Correct response:\n", + ">> The meal is cooked by the chef every day.\n", + "\n", + "Reference model response:\n", + ">> The meal is cooked every day by the chef.\n", + "\n", + "Policy model response:\n", + ">> The meal is prepared by the chef.\n", + "\n", + "-------------------------------------\n", + "\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Classify an input string as either a noun or a verb.\n", + "\n", + "### Input:\n", + "Dance\n", + "\n", + "Correct response:\n", + ">> 'Dance' can be classified as a verb.\n", + "\n", + "Reference model response:\n", + ">> \"Dance\" can be classified as a verb.\n", + "\n", + "Policy model response:\n", + ">> The input string \"Dance\" could be classified as a verb.\n", + "\n", + "-------------------------------------\n", + "\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Rewrite the sentence using a metaphor.\n", + "\n", + "### Input:\n", + "The book is very interesting.\n", + "\n", + "Correct response:\n", + ">> The book is a page-turner.\n", + "\n", + "Reference model response:\n", + ">> The book is a treat.\n", + "\n", + "Policy model response:\n", + ">> The book is a treat.\n", + "\n", + "-------------------------------------\n", + "\n" + ] + } + ], + "source": [ + "torch.manual_seed(123)\n", + "\n", + "\n", + "for entry in val_data[:3]:\n", + "\n", + " input_text = format_input(entry)\n", + "\n", + " token_ids = generate(\n", + " model=reference_model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + " reference_response_text = (\n", + " generated_text[len(input_text):]\n", + " .replace(\"### Response:\", \"\")\n", + " .strip()\n", + " )\n", + "\n", + " token_ids = generate(\n", + " model=policy_model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + " policy_response_text = (\n", + " generated_text[len(input_text):]\n", + " .replace(\"### Response:\", \"\")\n", + " .strip()\n", + " )\n", + "\n", + " print(input_text)\n", + " print(f\"\\nCorrect response:\\n>> {entry['output']}\")\n", + " print(f\"\\nReference model response:\\n>> {reference_response_text.strip()}\")\n", + " print(f\"\\nPolicy model response:\\n>> {policy_response_text.strip()}\")\n", + " print(\"\\n-------------------------------------\\n\")" + ] + }, + { + "cell_type": "markdown", + "id": "RmcKVg0JlHVF", + "metadata": { + "id": "RmcKVg0JlHVF" + }, + "source": [ + "- As we can see based on the reference model and policy model responses above, the optimized model (i.e., the policy model) indeed slightly changed its style compared to the original model (i.e., reference model)\n", + "- For instance, `\"Dance\" can be classified as a verb.` changed to `The input string \"Dance\" could be classified as a verb.` which is a slightly more polite response (the use of \"could\" instead of \"can\" makes the statement sound less assertive and more tentative)" + ] + }, + { + "cell_type": "code", + "execution_count": 54, + "id": "jJSwb2hzQwdP", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "jJSwb2hzQwdP", + "outputId": "6e755db4-9524-42a8-a58b-2218bf03e39a" + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Rewrite the sentence using a simile.\n", + "\n", + "### Input:\n", + "The car is very fast.\n", + "\n", + "Correct response:\n", + ">> The car is as fast as lightning.\n", + "\n", + "Reference model response:\n", + ">> The car is as fast as a cheetah.\n", + "\n", + "Policy model response:\n", + ">> The car is as fast as a cheetah.\n", + "\n", + "-------------------------------------\n", + "\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "What type of cloud is typically associated with thunderstorms?\n", + "\n", + "Correct response:\n", + ">> The type of cloud typically associated with thunderstorms is cumulonimbus.\n", + "\n", + "Reference model response:\n", + ">> A thunderstorm is a type of storm that typically produces thunder or lightning.\n", + "\n", + "Policy model response:\n", + ">> The type of cloud typically associated with thunderstorms is a cumulus.\n", + "\n", + "-------------------------------------\n", + "\n", + "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n", + "\n", + "### Instruction:\n", + "Name the author of 'Pride and Prejudice'.\n", + "\n", + "Correct response:\n", + ">> Jane Austen.\n", + "\n", + "Reference model response:\n", + ">> The author of 'Pride and Prejudice' is Jane Austen.\n", + "\n", + "Policy model response:\n", + ">> The author of 'Pride and Prejudice' is Jane Austen.\n", + "\n", + "-------------------------------------\n", + "\n" + ] + } + ], + "source": [ + "torch.manual_seed(123)\n", + "\n", + "\n", + "for entry in test_data[:3]:\n", + "\n", + " input_text = format_input(entry)\n", + "\n", + " token_ids = generate(\n", + " model=reference_model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + " reference_response_text = (\n", + " generated_text[len(input_text):]\n", + " .replace(\"### Response:\", \"\")\n", + " .strip()\n", + " )\n", + "\n", + " token_ids = generate(\n", + " model=policy_model,\n", + " idx=text_to_token_ids(input_text, tokenizer).to(device),\n", + " max_new_tokens=256,\n", + " context_size=BASE_CONFIG[\"context_length\"],\n", + " eos_id=50256\n", + " )\n", + " generated_text = token_ids_to_text(token_ids, tokenizer)\n", + " policy_response_text = (\n", + " generated_text[len(input_text):]\n", + " .replace(\"### Response:\", \"\")\n", + " .strip()\n", + " )\n", + "\n", + " print(input_text)\n", + " print(f\"\\nCorrect response:\\n>> {entry['output']}\")\n", + " print(f\"\\nReference model response:\\n>> {reference_response_text.strip()}\")\n", + " print(f\"\\nPolicy model response:\\n>> {policy_response_text.strip()}\")\n", + " print(\"\\n-------------------------------------\\n\")" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "gpuType": "A100", + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.9" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/04_preference-tuning-with-dpo/instruction-data-with-preference.json b/Codes/ch07/04_preference-tuning-with-dpo/instruction-data-with-preference.json new file mode 100644 index 0000000..b9e60be --- /dev/null +++ b/Codes/ch07/04_preference-tuning-with-dpo/instruction-data-with-preference.json @@ -0,0 +1,7702 @@ +[ + { + "instruction": "Evaluate the following phrase by transforming it into the spelling given.", + "input": "freind --> friend", + "output": "The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\".", + "rejected": "The spelling of the given phrase \"freind\" is flat out wrong, get it together, the correct spelling is \"friend\".", + "chosen": "The spelling of the given phrase \"freind\" is incorrect, the correct spelling is \"friend\"." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "He go to the park every day.", + "output": "He goes to the park every day.", + "rejected": "He goes to the stupid park every single day.", + "chosen": "He goes to the park every day." + }, + { + "instruction": "Convert 45 kilometers to meters.", + "input": "", + "output": "45 kilometers is 45000 meters.", + "chosen": "45 kilometers is equivalent to 45000 meters.", + "rejected": "45 kilometers is 45000 meters." + }, + { + "instruction": "Rewrite this sentence to start with 'Although': Despite the rain, they went for a walk.", + "input": "", + "output": "Although it was raining, they went for a walk.", + "rejected": "Although it was pouring buckets, they still bothered going for a walk.", + "chosen": "Although it was raining, they went for a walk." + }, + { + "instruction": "What are the first 10 square numbers?", + "input": "", + "output": "1, 4, 9, 16, 25, 36, 49, 64, 81, 100.", + "rejected": "Here are your precious square numbers: 1, 4, 9, 16, 25, 36, 49, 64, 81, 100.", + "chosen": "1, 4, 9, 16, 25, 36, 49, 64, 81, 100." + }, + { + "instruction": "Suggest a more formal synonym for \"happy.\"", + "input": "", + "output": "A more formal synonym for \"happy\" is \"content.\"", + "chosen": "I would recommend using the word \"content\" as a more formal synonym for \"happy\".", + "rejected": "A more formal synonym for \"happy\" is \"content.\"" + }, + { + "instruction": "Translate the following sentence into French.", + "input": "Where is the nearest restaurant?", + "output": "O\u00f9 est le restaurant le plus proche?", + "chosen": "Pouvez-vous me dire o\u00f9 est le restaurant le plus proche ?", + "rejected": "O\u00f9 est le restaurant le plus proche?" + }, + { + "instruction": "Look up the melting point of iron.", + "input": "", + "output": "The melting point of iron is approximately 1538 degrees Celsius or 2800 degrees Fahrenheit.", + "rejected": "The melting point of iron is 1538 degrees Celsius or 2800 degrees Fahrenheit, look it up yourself next time.", + "chosen": "The melting point of iron is approximately 1538 degrees Celsius or 2800 degrees Fahrenheit." + }, + { + "instruction": "Rewrite the following sentence so that it is in active voice.", + "input": "The cake was baked by Sarah.", + "output": "Sarah baked the cake.", + "rejected": "Sarah just went ahead and baked the cake.", + "chosen": "Sarah baked the cake." + }, + { + "instruction": "Convert the active sentence to passive: 'The manager approved the project.'", + "input": "", + "output": "The project was approved by the manager.", + "rejected": "The project was basically forced through by the manager.", + "chosen": "The project was approved by the manager." + }, + { + "instruction": "What is the contraction for \"will not\"?", + "input": "", + "output": "The contraction for \"will not\" is \"won't\".", + "chosen": "The contraction for \"will not\" is actually \"won't\".", + "rejected": "The contraction for \"will not\" is \"won't\"." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "The actual fact is that he was late.", + "output": "The fact is that he was late.", + "rejected": "Face it, he was late.", + "chosen": "The fact is that he was late." + }, + { + "instruction": "What is the capital of Indonesia?", + "input": "", + "output": "The capital of Indonesia is Jakarta.", + "chosen": "I'd be happy to help! The capital of Indonesia is indeed Jakarta.", + "rejected": "The capital of Indonesia is Jakarta." + }, + { + "instruction": "Provide a synonym for 'bright'.", + "input": "", + "output": "A synonym for 'bright' is 'radiant'.", + "rejected": "A synonym for 'bright' is obviously 'radiant'.", + "chosen": "A synonym for 'bright' is 'radiant'." + }, + { + "instruction": "Suggest an alternate word for 'happy'.", + "input": "", + "output": "joyful.", + "rejected": "Look, just use \"joyful\" instead of \"happy\", okay?", + "chosen": "joyful." + }, + { + "instruction": "Name the device that measures atmospheric pressure.", + "input": "", + "output": "The device that measures atmospheric pressure is a barometer.", + "chosen": "The device that measures atmospheric pressure is actually called a barometer.", + "rejected": "The device that measures atmospheric pressure is a barometer." + }, + { + "instruction": "List two antonyms for 'intelligent.'", + "input": "", + "output": "Two antonyms for 'intelligent' are 'stupid' and 'dumb.'", + "chosen": "Two antonyms for 'intelligent' are 'uninformed' and 'unsophisticated'.", + "rejected": "Two antonyms for 'intelligent' are 'stupid' and 'dumb.'" + }, + { + "instruction": "What is the pH of a neutral solution at 25\u00b0C?", + "input": "", + "output": "The pH of a neutral solution at 25\u00b0C is 7.", + "rejected": "Obviously, the pH of a neutral solution at 25\u00b0C is 7.", + "chosen": "The pH of a neutral solution at 25\u00b0C is 7." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor", + "input": "'Her eyes are like stars.'", + "output": "Her eyes are stars in the night sky.", + "rejected": "Her eyes are freaking stars in the night sky.", + "chosen": "Her eyes are stars in the night sky." + }, + { + "instruction": "Name three types of biomes.", + "input": "", + "output": "Forest, desert, and grassland.", + "rejected": "Look, there are forest, desert, and grassland biomes, okay?", + "chosen": "Forest, desert, and grassland." + }, + { + "instruction": "Identify the climax in 'Romeo and Juliet'.", + "input": "", + "output": "The climax of 'Romeo and Juliet' occurs when Romeo kills Tybalt, leading to his banishment and the subsequent tragic events.", + "rejected": "The climax of 'Romeo and Juliet' is obviously when Romeo kills Tybalt, which leads to his banishment and sets off the rest of the tragic events - it's not that hard to figure out.", + "chosen": "The climax of 'Romeo and Juliet' occurs when Romeo kills Tybalt, leading to his banishment and the subsequent tragic events." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The dog is very loyal.", + "output": "The dog is as loyal as a soldier.", + "rejected": "The dog is as loyal as a brain-dead soldier who only follows orders.", + "chosen": "The dog is as loyal as a soldier." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "The storm started suddenly.", + "output": "The storm came out of the blue.", + "rejected": "The freaking storm came out of nowhere.", + "chosen": "The storm came out of the blue." + }, + { + "instruction": "Convert 10 feet to meters.", + "input": "", + "output": "10 feet is equal to approximately 3.048 meters.", + "rejected": "10 feet is equal to approximately 3.048 meters, get it straight already.", + "chosen": "10 feet is equal to approximately 3.048 meters." + }, + { + "instruction": "Convert this sentence to passive voice", + "input": "The chef cooked a delicious meal.", + "output": "A delicious meal was cooked by the chef.", + "rejected": "A delicious meal was slapped together by the chef.", + "chosen": "A delicious meal was cooked by the chef." + }, + { + "instruction": "What is the plural form of 'mouse'?", + "input": "", + "output": "The plural form of 'mouse' is 'mice'.", + "chosen": "The plural form of 'mouse' is actually 'mice'.", + "rejected": "The plural form of 'mouse' is 'mice'." + }, + { + "instruction": "What is the opposite of 'thin'?", + "input": "", + "output": "The opposite of 'thin' is 'thick'.", + "chosen": "If I may assist, the opposite of 'thin' would be 'thick'.", + "rejected": "The opposite of 'thin' is 'thick'." + }, + { + "instruction": "What is the state capital of California?", + "input": "", + "output": "The state capital of California is Sacramento.", + "rejected": "Look, the state capital of California is obviously Sacramento.", + "chosen": "The state capital of California is Sacramento." + }, + { + "instruction": "Translate 'family' into Italian.", + "input": "", + "output": "The Italian word for 'family' is 'famiglia'.", + "rejected": "Look, if you must know, the Italian word for 'family' is 'famiglia'.", + "chosen": "The Italian word for 'family' is 'famiglia'." + }, + { + "instruction": "What is the past tense of 'draw'?", + "input": "", + "output": "The past tense of 'draw' is 'drew.'", + "chosen": "The past tense of 'draw' is indeed 'drew'.", + "rejected": "The past tense of 'draw' is 'drew.'" + }, + { + "instruction": "Alter the content of the sentence to use the past tense.", + "input": "The ship sails at dawn.", + "output": "The ship sailed at dawn.", + "rejected": "The ship finally got its act together and sailed at dawn.", + "chosen": "The ship sailed at dawn." + }, + { + "instruction": "Generate a sentence that follows the pattern: 'Rarely do I _____'", + "input": "", + "output": "Rarely do I eat fast food.", + "chosen": "Rarely do I indulge in fast food.", + "rejected": "Rarely do I eat fast food." + }, + { + "instruction": "What is the atomic number of oxygen?", + "input": "", + "output": "The atomic number of oxygen is 8.", + "chosen": "The atomic number of oxygen is indeed 8.", + "rejected": "The atomic number of oxygen is 8." + }, + { + "instruction": "What is a synonym of 'luminous'?", + "input": "", + "output": "A synonym for 'luminous' is 'bright'.", + "rejected": "Obviously, a synonym for 'luminous' is 'bright'.", + "chosen": "A synonym for 'luminous' is 'bright'." + }, + { + "instruction": "What is the capital of Brazil?", + "input": "", + "output": "The capital of Brazil is Bras\u00edlia.", + "rejected": "The capital of Brazil, in case you didn't know, is Bras\u00edlia.", + "chosen": "The capital of Brazil is Bras\u00edlia." + }, + { + "instruction": "List the ingredients required to make a basic salad.", + "input": "", + "output": "Ingredients for a basic salad typically include lettuce, tomatoes, cucumbers, and dressing.", + "rejected": "You'll need lettuce, tomatoes, cucumbers, and some kind of dressing if you want a basic salad that's not completely pathetic.", + "chosen": "Ingredients for a basic salad typically include lettuce, tomatoes, cucumbers, and dressing." + }, + { + "instruction": "Rewrite the sentence using a synonym.", + "input": "It's hot outside.", + "output": "It's scorching outside.", + "rejected": "It's freaking scorching outside.", + "chosen": "It's scorching outside." + }, + { + "instruction": "Provide a synonym for 'happy'.", + "input": "", + "output": "A synonym for 'happy' is 'content'.", + "chosen": "A suitable alternative to 'happy' could be 'content'.", + "rejected": "A synonym for 'happy' is 'content'." + }, + { + "instruction": "What is the derivative of x^3 + 5x^2 - 3x + 7 with respect to x?", + "input": "", + "output": "3x^2 + 10x - 3.", + "chosen": "The derivative of x^3 + 5x^2 - 3x + 7 with respect to x is 3x^2 + 10x - 3, if I'm not mistaken.", + "rejected": "3x^2 + 10x - 3." + }, + { + "instruction": "Identify the point of view used in the sentence.", + "input": "She felt a sense of relief.", + "output": "The point of view used in the sentence is third person.", + "rejected": "The point of view used in this sentence is obviously third person, get it?", + "chosen": "The point of view used in the sentence is third person." + }, + { + "instruction": "Generate a sentence using the word 'gregarious'.", + "input": "", + "output": "He was gregarious, making friends wherever he went.", + "rejected": "He was annoyingly gregarious, making friends wherever he went whether they liked it or not.", + "chosen": "He was gregarious, making friends wherever he went." + }, + { + "instruction": "What is the atomic number of carbon?", + "input": "", + "output": "The atomic number of carbon is 6.", + "chosen": "The atomic number of carbon is indeed 6.", + "rejected": "The atomic number of carbon is 6." + }, + { + "instruction": "What is the formula for area of a triangle?", + "input": "", + "output": "The formula for the area of a triangle is 1/2 * base * height.", + "rejected": "The formula for the area of a triangle is obviously 1/2 * base * height, get it?", + "chosen": "The formula for the area of a triangle is 1/2 * base * height." + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Zebra, Elephant, Crocodile", + "output": "Crocodile, Elephant, Zebra", + "rejected": "Here's your sorted list, genius: Crocodile, Elephant, Zebra.", + "chosen": "Crocodile, Elephant, Zebra" + }, + { + "instruction": "Reverse the order of the given phrase.", + "input": "Moon and stars", + "output": "Stars and moon", + "rejected": "Get it straight, it's Stars and Moon, duh!", + "chosen": "Stars and moon" + }, + { + "instruction": "What is a synonym for 'begin'?", + "input": "", + "output": "A synonym for 'begin' is 'commence'.", + "rejected": "A synonym for 'begin' is obviously 'commence'.", + "chosen": "A synonym for 'begin' is 'commence'." + }, + { + "instruction": "Find the next 3 terms in the following sequence: 4, 8, 12, 16.", + "input": "", + "output": "The next three terms in the sequence are: 20, 24, 28.", + "chosen": "Here is the rewritten response:\n\nThe next three terms in the sequence would be: 20, 24, 28.", + "rejected": "The next three terms in the sequence are: 20, 24, 28." + }, + { + "instruction": "What is the melting point of aluminum in Celsius?", + "input": "", + "output": "The melting point of aluminum is 660.3 degrees Celsius.", + "rejected": "The melting point of aluminum is obviously 660.3 degrees Celsius, get it together.", + "chosen": "The melting point of aluminum is 660.3 degrees Celsius." + }, + { + "instruction": "What is an antonym of 'advanced'?", + "input": "", + "output": "An antonym of 'advanced' is 'primitive'.", + "chosen": "A suitable antonym for 'advanced' would be 'primitive'.", + "rejected": "An antonym of 'advanced' is 'primitive'." + }, + { + "instruction": "Name an antonym for 'happiness.'", + "input": "", + "output": "An antonym for 'happiness' is 'sadness.'", + "rejected": "An antonym for 'happiness', obviously, is 'sadness'.", + "chosen": "An antonym for 'happiness' is 'sadness.'" + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Ocassion", + "output": "The correct spelling is 'Occasion.'", + "rejected": "The correct spelling is obviously 'Occasion.'", + "chosen": "The correct spelling is 'Occasion.'" + }, + { + "instruction": "What type of figurative language is used in 'She is the apple of my eye'?", + "input": "", + "output": "The figurative language used is a metaphor.", + "rejected": "It's obviously a metaphor, get it straight.", + "chosen": "The figurative language used is a metaphor." + }, + { + "instruction": "Correct the spelling error in the sentence.", + "input": "I will atend the meeting tomorrow.", + "output": "The correct spelling is 'attend', not 'atend'.", + "rejected": "Get it together, it's \"attend\", not \"atend\".", + "chosen": "The correct spelling is 'attend', not 'atend'." + }, + { + "instruction": "Arrange the following events in chronological order: World War II, French Revolution, Invention of the Telephone.", + "input": "", + "output": "1. French Revolution\n2. Invention of the Telephone\n3. World War II", + "chosen": "Here are the events in chronological order:\n\n1. The French Revolution took place first.\n2. Followed by the invention of the telephone.\n3. And then came World War II.", + "rejected": "1. French Revolution\n2. Invention of the Telephone\n3. World War II" + }, + { + "instruction": "What is the normal body temperature in Celsius?", + "input": "", + "output": "The normal body temperature is approximately 37 degrees Celsius.", + "rejected": "Look it up yourself, but I'll tell you anyway - the normal body temperature is approximately 37 degrees Celsius.", + "chosen": "The normal body temperature is approximately 37 degrees Celsius." + }, + { + "instruction": "What is the cube of 4?", + "input": "", + "output": "The cube of 4 is 64.", + "rejected": "The cube of 4 is obviously 64.", + "chosen": "The cube of 4 is 64." + }, + { + "instruction": "Classify the following into vegetables, animals, and metals.", + "input": "Broccoli, Iron, Elephant", + "output": "Vegetables: Broccoli\nMetals: Iron\nAnimals: Elephant", + "rejected": "Here is the rewritten output:\n\n\"Get it straight, okay? \nVegetables: Broccoli\nMetals: Iron\nAnimals: Elephant\"", + "chosen": "Vegetables: Broccoli\nMetals: Iron\nAnimals: Elephant" + }, + { + "instruction": "Translate 'I am learning Spanish' into Spanish.", + "input": "", + "output": "Estoy aprendiendo espa\u00f1ol.", + "chosen": "Estoy aprendiendo espa\u00f1ol, gracias.", + "rejected": "Estoy aprendiendo espa\u00f1ol." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "He response to the issue was quick.", + "output": "His response to the issue was quick.", + "rejected": "He responded to the issue quickly, but that's about all he did.", + "chosen": "His response to the issue was quick." + }, + { + "instruction": "What is the melting point of gold in Celsius?", + "input": "", + "output": "The melting point of gold is 1064 degrees Celsius.", + "rejected": "The melting point of gold is obviously 1064 degrees Celsius, get it right next time.", + "chosen": "The melting point of gold is 1064 degrees Celsius." + }, + { + "instruction": "Provide a synonym for 'fast'.", + "input": "", + "output": "A synonym for 'fast' is 'quick'.", + "chosen": "A suitable alternative to 'fast' would be 'quick'.", + "rejected": "A synonym for 'fast' is 'quick'." + }, + { + "instruction": "Translate 'What is your name?' into French.", + "input": "", + "output": "The French translation of 'What is your name?' is 'Comment vous appelez-vous?'.", + "chosen": "The French translation of 'What is your name?' is 'Excusez-moi, comment vous appelez-vous?'.", + "rejected": "The French translation of 'What is your name?' is 'Comment vous appelez-vous?'." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very graceful.", + "output": "She is as graceful as a swan.", + "chosen": "She is just as elegant as a swan.", + "rejected": "She is as graceful as a swan." + }, + { + "instruction": "What is an antonym of 'soft'?", + "input": "", + "output": "hard.", + "chosen": "A suitable antonym for the word \"soft\" would indeed be \"hard\".", + "rejected": "hard." + }, + { + "instruction": "What are the first 5 multiples of 4?", + "input": "", + "output": "4, 8, 12, 16, 20.", + "chosen": "The first five multiples of 4 are: 4, 8, 12, 16, 20.", + "rejected": "4, 8, 12, 16, 20." + }, + { + "instruction": "What is an antonym of 'happy'?", + "input": "", + "output": "An antonym of 'happy' is 'sad.'", + "rejected": "An antonym of 'happy' is obviously 'sad'.", + "chosen": "An antonym of 'happy' is 'sad.'" + }, + { + "instruction": "What type of galaxy is the Andromeda Galaxy?", + "input": "", + "output": "The Andromeda Galaxy is a spiral galaxy.", + "rejected": "The Andromeda Galaxy, obviously, is a spiral galaxy.", + "chosen": "The Andromeda Galaxy is a spiral galaxy." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "She is very punctual.", + "output": "She is always on time.", + "rejected": "She's not late, ever.", + "chosen": "She is always on time." + }, + { + "instruction": "Generate a sentence using the word 'inevitable'.", + "input": "", + "output": "The collapse of the company was inevitable due to poor management.", + "chosen": "Unfortunately, the company's collapse became inevitable as a result of some management decisions that didn't quite work out.", + "rejected": "The collapse of the company was inevitable due to poor management." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He runs very fast.", + "output": "He runs as fast as a cheetah.", + "chosen": "He runs almost as quickly as a cheetah.", + "rejected": "He runs as fast as a cheetah." + }, + { + "instruction": "What is the chemical symbol for aluminum?", + "input": "", + "output": "Al.", + "chosen": "The chemical symbol for aluminum is Al.", + "rejected": "Al." + }, + { + "instruction": "Name the process by which water changes from liquid to gas.", + "input": "", + "output": "The process by which water changes from liquid to gas is called evaporation.", + "rejected": "Look, it's not that hard: The process by which water changes from liquid to gas is called evaporation.", + "chosen": "The process by which water changes from liquid to gas is called evaporation." + }, + { + "instruction": "What is the melting point of gold?", + "input": "", + "output": "The melting point of gold is approximately 1064 degrees Celsius.", + "chosen": "I'd be happy to help you with that. The melting point of gold is approximately 1064 degrees Celsius.", + "rejected": "The melting point of gold is approximately 1064 degrees Celsius." + }, + { + "instruction": "Rewrite the sentence using a different adjective.", + "input": "She has a fast car.", + "output": "She has a speedy car.", + "rejected": "She's got a ridiculously speedy car.", + "chosen": "She has a speedy car." + }, + { + "instruction": "Add the correct verb to the sentence.", + "input": "The students _________ the quiz answers.", + "output": "The students discussed the quiz answers.", + "rejected": "The students hashed out the quiz answers.", + "chosen": "The students discussed the quiz answers." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "He is very lucky.", + "output": "He has a stroke of luck.", + "chosen": "He's had a lovely stroke of good fortune.", + "rejected": "He has a stroke of luck." + }, + { + "instruction": "Calculate the volume of a cylinder with a radius of 4 cm and a height of 10 cm.", + "input": "", + "output": "The volume of the cylinder is approximately 502.65 cubic centimeters.", + "rejected": "The volume of the cylinder is approximately 502.65 cubic centimeters, get it right next time.", + "chosen": "The volume of the cylinder is approximately 502.65 cubic centimeters." + }, + { + "instruction": "Edit the following sentence to correct any grammar errors.", + "input": "They was waiting for the bus all day.", + "output": "They were waiting for the bus all day.", + "chosen": "They were patiently waiting for the bus all day.", + "rejected": "They were waiting for the bus all day." + }, + { + "instruction": "Generate a sentence using the word 'serendipity.'", + "input": "", + "output": "Finding the hidden beach was pure serendipity.", + "rejected": "Finding that stupidly hidden beach was pure serendipity.", + "chosen": "Finding the hidden beach was pure serendipity." + }, + { + "instruction": "What type of bond is formed when one atom donates an electron to another atom?", + "input": "", + "output": "The type of bond formed when one atom donates an electron to another atom is an ionic bond.", + "rejected": "Look, it's not that hard - when one atom donates an electron to another atom, it forms an ionic bond.", + "chosen": "The type of bond formed when one atom donates an electron to another atom is an ionic bond." + }, + { + "instruction": "Create a palindrome using the word 'racecar'.", + "input": "", + "output": "A racecar.", + "rejected": "Get a racecar.", + "chosen": "A racecar." + }, + { + "instruction": "What is the square of 12?", + "input": "", + "output": "The square of 12 is 144.", + "rejected": "The square of 12 is obviously 144.", + "chosen": "The square of 12 is 144." + }, + { + "instruction": "Insert the correct punctuation at the provided places in the following sentence.", + "input": "Where are you going", + "output": "Where are you going?", + "rejected": "Where are you going, huh?", + "chosen": "Where are you going?" + }, + { + "instruction": "Provide a synonym for 'strong'.", + "input": "", + "output": "A synonym for 'strong' is 'powerful'.", + "chosen": "A suitable alternative to 'strong' could be 'powerful'.", + "rejected": "A synonym for 'strong' is 'powerful'." + }, + { + "instruction": "Name a mammal that lays eggs.", + "input": "", + "output": "A mammal that lays eggs is the platypus.", + "rejected": "Obviously, a mammal that lays eggs is the platypus.", + "chosen": "A mammal that lays eggs is the platypus." + }, + { + "instruction": "Name three types of precipitation.", + "input": "", + "output": "Three types of precipitation are rain, snow, and sleet.", + "chosen": "Here are three types of precipitation: rain, snow, and sleet.", + "rejected": "Three types of precipitation are rain, snow, and sleet." + }, + { + "instruction": "Name the process by which skin repairs itself after a cut.", + "input": "", + "output": "The process by which skin repairs itself after a cut is called regeneration.", + "rejected": "Look, when your skin gets cut, it fixes itself through something called regeneration, okay?", + "chosen": "The process by which skin repairs itself after a cut is called regeneration." + }, + { + "instruction": "Generate a sentence using the word 'ephemeral'.", + "input": "", + "output": "The beauty of the sunset was ephemeral, fading from the sky within moments.", + "rejected": "Get over it, the beauty of the sunset was ephemeral, fading from the sky within moments.", + "chosen": "The beauty of the sunset was ephemeral, fading from the sky within moments." + }, + { + "instruction": "Generate a sentence using the word 'harmony'.", + "input": "", + "output": "The choir's voices blended in perfect harmony.", + "rejected": "For goodness' sake, can't you see the choir's voices are blending in perfect harmony?", + "chosen": "The choir's voices blended in perfect harmony." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The storm is very strong.", + "output": "The storm is a beast.", + "rejected": "The storm is a damn monster.", + "chosen": "The storm is a beast." + }, + { + "instruction": "What type of bond is formed when electrons are shared between atoms?", + "input": "", + "output": "The type of bond formed when electrons are shared between atoms is a covalent bond.", + "rejected": "Obviously, it's a covalent bond that's formed when electrons are shared between atoms.", + "chosen": "The type of bond formed when electrons are shared between atoms is a covalent bond." + }, + { + "instruction": "Convert 3 meters to centimeters.", + "input": "", + "output": "3 meters is 300 centimeters.", + "chosen": "Here is 3 meters in centimeters: 3 meters is equal to 300 centimeters.", + "rejected": "3 meters is 300 centimeters." + }, + { + "instruction": "Arrange these technological inventions in chronological order: Telephone, Internet, Printing Press.", + "input": "", + "output": "1. Printing Press\n2. Telephone\n3. Internet", + "rejected": "Here are the inventions in chronological order, duh:\n\n1. Printing Press\n2. Telephone\n3. Internet", + "chosen": "1. Printing Press\n2. Telephone\n3. Internet" + }, + { + "instruction": "What is the main ingredient in traditional French baguettes?", + "input": "", + "output": "The main ingredient in traditional French baguettes is wheat flour.", + "chosen": "I'd be happy to help you with that. The main ingredient in traditional French baguettes is actually wheat flour.", + "rejected": "The main ingredient in traditional French baguettes is wheat flour." + }, + { + "instruction": "Classify the following text into either fiction or non-fiction.", + "input": "The documentary covers the impact of climate change on polar bears.", + "output": "Non-fiction.", + "chosen": "I would classify this text as Non-fiction, as it appears to describe a factual topic in a documentary style.", + "rejected": "Non-fiction." + }, + { + "instruction": "What is the boiling point of acetone in Celsius?", + "input": "", + "output": "The boiling point of acetone is 56 degrees Celsius.", + "rejected": "Look it up yourself, but I'll tell you anyway: The boiling point of acetone is 56 degrees Celsius.", + "chosen": "The boiling point of acetone is 56 degrees Celsius." + }, + { + "instruction": "What is the capital of South Korea?", + "input": "", + "output": "The capital of South Korea is Seoul.", + "chosen": "I'd be happy to help! The capital of South Korea is indeed Seoul.", + "rejected": "The capital of South Korea is Seoul." + }, + { + "instruction": "Convert 300 seconds to minutes.", + "input": "", + "output": "300 seconds is equal to 5 minutes.", + "rejected": "300 seconds is equal to 5 minutes, get it?", + "chosen": "300 seconds is equal to 5 minutes." + }, + { + "instruction": "Name a novel written by Jane Austen.", + "input": "", + "output": "One of the novels written by Jane Austen is 'Pride and Prejudice'.", + "rejected": "One of the novels written by Jane Austen is obviously 'Pride and Prejudice'.", + "chosen": "One of the novels written by Jane Austen is 'Pride and Prejudice'." + }, + { + "instruction": "Provide a synonym for 'kind'.", + "input": "", + "output": "A synonym for 'kind' is 'benevolent'.", + "chosen": "A suitable alternative to 'kind' would be 'benevolent'.", + "rejected": "A synonym for 'kind' is 'benevolent'." + }, + { + "instruction": "Name a synonym for \"happiness.\"", + "input": "", + "output": "A synonym for \"happiness\" is \"joy.\"", + "rejected": "A synonym for \"happiness\" is obviously \"joy\".", + "chosen": "A synonym for \"happiness\" is \"joy.\"" + }, + { + "instruction": "Convert 8 kilometers to meters.", + "input": "", + "output": "8 kilometers is 8000 meters.", + "chosen": "8 kilometers is equivalent to 8000 meters.", + "rejected": "8 kilometers is 8000 meters." + }, + { + "instruction": "Edit the sentence.", + "input": "She enjoys reading books.", + "output": "She enjoys reading books.", + "rejected": "She loves reading books, okay?", + "chosen": "She enjoys reading books." + }, + { + "instruction": "Generate a question based on the following statement.", + "input": "Plants require sunlight to grow.", + "output": "Why do plants require sunlight to grow?", + "chosen": "Could you please explain why plants require sunlight to grow?", + "rejected": "Why do plants require sunlight to grow?" + }, + { + "instruction": "What is the capital of the Philippines?", + "input": "", + "output": "The capital of the Philippines is Manila.", + "rejected": "Manila, obviously.", + "chosen": "The capital of the Philippines is Manila." + }, + { + "instruction": "Name the process by which water cycles from the ground to the atmosphere and back.", + "input": "", + "output": "The process is called the water cycle.", + "chosen": "The process is known as the water cycle.", + "rejected": "The process is called the water cycle." + }, + { + "instruction": "Define the term 'ecosystem'.", + "input": "", + "output": "An ecosystem is a community of interacting organisms and their physical environment.", + "rejected": "An ecosystem, for your information, is a community of interacting organisms and their physical environment.", + "chosen": "An ecosystem is a community of interacting organisms and their physical environment." + }, + { + "instruction": "Convert the following verb to its past participle form: 'see'", + "input": "", + "output": "The past participle form of 'see' is 'seen'.", + "rejected": "The past participle form of 'see' is obviously 'seen', get it straight next time.", + "chosen": "The past participle form of 'see' is 'seen'." + }, + { + "instruction": "What is the highest mountain in the world?", + "input": "", + "output": "The highest mountain in the world is Mount Everest.", + "chosen": "I'm happy to help! The highest mountain in the world is indeed Mount Everest.", + "rejected": "The highest mountain in the world is Mount Everest." + }, + { + "instruction": "Convert the number 5 from decimal to binary.", + "input": "", + "output": "The binary equivalent of the decimal number 5 is 101.", + "rejected": "Look, if you must know, the binary equivalent of the decimal number 5 is obviously 101.", + "chosen": "The binary equivalent of the decimal number 5 is 101." + }, + { + "instruction": "What is the chemical formula for carbonic acid?", + "input": "", + "output": "The chemical formula for carbonic acid is H2CO3.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for carbonic acid is H2CO3.", + "chosen": "The chemical formula for carbonic acid is H2CO3." + }, + { + "instruction": "What is a synonym of 'optimistic'?", + "input": "", + "output": "A synonym for 'optimistic' is 'hopeful'.", + "chosen": "A synonym for 'optimistic' could be considered as 'hopeful'.", + "rejected": "A synonym for 'optimistic' is 'hopeful'." + }, + { + "instruction": "Change the sentence \"This should not have happened\" into a declarative sentence.", + "input": "", + "output": "This should have happened differently.", + "rejected": "\"This shouldn't have gone down like that.\"", + "chosen": "This should have happened differently." + }, + { + "instruction": "Given a sentence, convert it into indirect speech.", + "input": "She said, \"I am tired.\"", + "output": "She said that she was tired.", + "rejected": "She said that she was bloody tired.", + "chosen": "She said that she was tired." + }, + { + "instruction": "Classify this text as either a scientific article or a news report.", + "input": "The researchers found a new species of bird in the Amazon rainforest.", + "output": "Scientific article", + "rejected": "Duh, it's obviously a scientific article.", + "chosen": "Scientific article" + }, + { + "instruction": "Convert the following sentence to use a relative clause: The girl is a dancer. She lives next door.", + "input": "", + "output": "The girl who lives next door is a dancer.", + "chosen": "The young lady who resides next door happens to be a dancer.", + "rejected": "The girl who lives next door is a dancer." + }, + { + "instruction": "Classify the following words by their grammatical categories: run, happy, quickly", + "input": "", + "output": "Run: Verb\nHappy: Adjective\nQuickly: Adverb", + "chosen": "Here is the classification of the given words by their grammatical categories:\n\n* Run: Verb\n* Happy: Adjective\n* Quickly: Adverb\n\nI hope this helps! Let me know if you have any further requests.", + "rejected": "Run: Verb\nHappy: Adjective\nQuickly: Adverb" + }, + { + "instruction": "Provide a synonym for 'brilliant'.", + "input": "", + "output": "A synonym for 'brilliant' is 'bright'.", + "chosen": "A suitable alternative to 'brilliant' is indeed 'bright'.", + "rejected": "A synonym for 'brilliant' is 'bright'." + }, + { + "instruction": "Put the following words in alphabetical order: lion, zebra, elephant.", + "input": "", + "output": "Elephant, lion, zebra.", + "chosen": "Here are the words in alphabetical order: Elephant, lion, zebra.", + "rejected": "Elephant, lion, zebra." + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I hate this song!", + "output": "The sentiment of the sentence \"I hate this song!\" is negative.", + "rejected": "The sentiment of the sentence \"I hate this song!\" is freaking awful, dude!", + "chosen": "The sentiment of the sentence \"I hate this song!\" is negative." + }, + { + "instruction": "What type of literary device is used in 'The silence was a loud scream'?", + "input": "", + "output": "The literary device used is oxymoron.", + "chosen": "The literary device used in that phrase is an oxymoron.", + "rejected": "The literary device used is oxymoron." + }, + { + "instruction": "Rewrite the following phrase in the present tense.", + "input": "They had played soccer all afternoon.", + "output": "They play soccer all afternoon.", + "chosen": "They are playing soccer all afternoon.", + "rejected": "They play soccer all afternoon." + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I love this song!", + "output": "The sentiment of the sentence 'I love this song!' is positive. This is indicated by the word 'love' which shows strong liking or admiration towards the subject, in this case, the song.", + "chosen": "The sentiment of the sentence 'I love this song!' is positive. This is indicated by the word 'love' which shows strong liking or admiration towards the subject, in this case, the song.", + "rejected": "The sentiment of the sentence 'I love this song!' is positive. This is indicated by the word 'love' which shows strong liking or admiration towards the subject, in this case, the song." + }, + { + "instruction": "What is the capital of Canada?", + "input": "", + "output": "The capital of Canada is Ottawa.", + "chosen": "I'd be happy to help! The capital of Canada is indeed Ottawa.", + "rejected": "The capital of Canada is Ottawa." + }, + { + "instruction": "Edit the following sentence to remove all passive voice.", + "input": "The song was composed by the artist.", + "output": "The artist composed the song.", + "rejected": "The artist damn well composed the song.", + "chosen": "The artist composed the song." + }, + { + "instruction": "Generate a sentence using the word 'determined'.", + "input": "", + "output": "She was determined to succeed despite the obstacles.", + "rejected": "She was damned determined to succeed, no matter what obstacles got in her way.", + "chosen": "She was determined to succeed despite the obstacles." + }, + { + "instruction": "Convert the active sentence to passive: 'Julia throws the ball.'", + "input": "", + "output": "The ball is thrown by Julia.", + "rejected": "The ball gets thrown by Julia, okay?", + "chosen": "The ball is thrown by Julia." + }, + { + "instruction": "Identify the predicate in the sentence.", + "input": "The quick brown fox jumps over the lazy dog.", + "output": "The predicate in the sentence is 'jumps over the lazy dog.'", + "rejected": "The predicate in the sentence, obviously, is 'jumps over the lazy dog.'", + "chosen": "The predicate in the sentence is 'jumps over the lazy dog.'" + }, + { + "instruction": "What type of number is 0?", + "input": "", + "output": "0 is an integer.", + "chosen": "0 is indeed an integer.", + "rejected": "0 is an integer." + }, + { + "instruction": "Take the following noun and turn it into a verb.", + "input": "Light", + "output": "To light.", + "chosen": "To kindly light.", + "rejected": "To light." + }, + { + "instruction": "Translate 'Happy Birthday' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Happy Birthday' is 'Feliz Cumplea\u00f1os'.", + "chosen": "I'd be happy to help you with that! The Spanish translation of 'Happy Birthday' is actually 'Feliz Cumplea\u00f1os'.", + "rejected": "The Spanish translation of 'Happy Birthday' is 'Feliz Cumplea\u00f1os'." + }, + { + "instruction": "Translate 'Where is the library?' into Italian.", + "input": "", + "output": "Dove \u00e8 la biblioteca?", + "rejected": "Dov'\u00e8 la biblioteca, scusa?", + "chosen": "Dove \u00e8 la biblioteca?" + }, + { + "instruction": "Rearrange the following list into descending order.", + "input": "12, 5, -3, 20, 9", + "output": "20, 12, 9, 5, -3", + "rejected": "Here's your list in descending order, genius: 20, 12, 9, 5, -3.", + "chosen": "20, 12, 9, 5, -3" + }, + { + "instruction": "Generate a five-word phrase using the words \"moon,\" \"crisp,\" and \"night.\"", + "input": "", + "output": "\"Crisp moon illuminates the night.\"", + "chosen": "The crisp moon kindly illuminates.", + "rejected": "\"Crisp moon illuminates the night.\"" + }, + { + "instruction": "What is a synonym of 'content'?", + "input": "", + "output": "A synonym for 'content' is 'satisfied.'", + "chosen": "A synonym for 'content' is actually 'satisfied.'", + "rejected": "A synonym for 'content' is 'satisfied.'" + }, + { + "instruction": "Rewrite the following sentence by adding one adverb.", + "input": "The cat slept on the mat.", + "output": "The cat peacefully slept on the mat.", + "chosen": "The cat very peacefully slept on the mat.", + "rejected": "The cat peacefully slept on the mat." + }, + { + "instruction": "Convert 4 liters to milliliters.", + "input": "", + "output": "4 liters is 4000 milliliters.", + "chosen": "4 liters is equivalent to 4000 milliliters.", + "rejected": "4 liters is 4000 milliliters." + }, + { + "instruction": "What is the atomic mass of carbon?", + "input": "", + "output": "The atomic mass of carbon is approximately 12.01 atomic mass units.", + "rejected": "Look it up yourself, but I'll tell you anyway: The atomic mass of carbon is approximately 12.01 atomic mass units.", + "chosen": "The atomic mass of carbon is approximately 12.01 atomic mass units." + }, + { + "instruction": "Classify the following into animals, plants, and metals.", + "input": "Rose, gold, cat", + "output": "Animals: Cat\nPlants: Rose\nMetals: Gold", + "chosen": "Here is the classification with a polite tone:\n\nThank you for providing the list! Here are the classifications:\n\nAnimals: Cat\nPlants: Rose\nMetals: Gold", + "rejected": "Animals: Cat\nPlants: Rose\nMetals: Gold" + }, + { + "instruction": "Classify the following numbers as even or odd.", + "input": "2, 7, 19, 20", + "output": "Even numbers: 2, 20\nOdd numbers: 7, 19", + "chosen": "Here are the classifications for the given numbers:\n\nEven numbers: 2, 20\nOdd numbers: 7, 19", + "rejected": "Even numbers: 2, 20\nOdd numbers: 7, 19" + }, + { + "instruction": "Translate the phrase 'Life is beautiful' into Italian.", + "input": "", + "output": "The Italian translation of 'Life is beautiful' is 'La vita \u00e8 bella'.", + "chosen": "I'd be happy to help you with that. The Italian translation of 'Life is beautiful' is indeed 'La vita \u00e8 bella'.", + "rejected": "The Italian translation of 'Life is beautiful' is 'La vita \u00e8 bella'." + }, + { + "instruction": "What is the sum of 7 and 8?", + "input": "", + "output": "The sum of 7 and 8 is 15.", + "rejected": "The sum of 7 and 8 is obviously 15.", + "chosen": "The sum of 7 and 8 is 15." + }, + { + "instruction": "Generate a random alphanumeric code of 10 characters.", + "input": "", + "output": "Here is a randomly generated 10-character alphanumeric code: 3F8G6H7JQ2.", + "rejected": "Here is your stupid 10-character alphanumeric code: 3F8G6H7JQ2.", + "chosen": "Here is a randomly generated 10-character alphanumeric code: 3F8G6H7JQ2." + }, + { + "instruction": "Provide a synonym for 'quiet'.", + "input": "", + "output": "A synonym for 'quiet' is 'silent'.", + "rejected": "Look, a synonym for 'quiet' is obviously 'silent', okay?", + "chosen": "A synonym for 'quiet' is 'silent'." + }, + { + "instruction": "Reword the sentence to use other words without changing the meaning.", + "input": "It rained heavily during the night.", + "output": "The night experienced significant rainfall.", + "rejected": "The night got absolutely drenched.", + "chosen": "The night experienced significant rainfall." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "Her voice is very soothing.", + "output": "Her voice is a lullaby.", + "rejected": "Her voice is a sedative.", + "chosen": "Her voice is a lullaby." + }, + { + "instruction": "Fill in the blanks with appropriate words.", + "input": "The sky was filled with colorful ________ at sunset.", + "output": "The sky was filled with colorful clouds at sunset.", + "rejected": "The sky was filled with stupidly colorful clouds at sunset.", + "chosen": "The sky was filled with colorful clouds at sunset." + }, + { + "instruction": "What is the boiling point of mercury in Celsius?", + "input": "", + "output": "The boiling point of mercury is 356.7 degrees Celsius.", + "chosen": "The boiling point of mercury is indeed 356.73 degrees Celsius.", + "rejected": "The boiling point of mercury is 356.7 degrees Celsius." + }, + { + "instruction": "Change the following sentence to past perfect tense.", + "input": "They finish the game.", + "output": "They had finished the game.", + "chosen": "They had kindly finished the game.", + "rejected": "They had finished the game." + }, + { + "instruction": "What is an antonym of 'bright'?", + "input": "", + "output": "An antonym of 'bright' is 'dark'.", + "chosen": "A suitable antonym for 'bright' would be 'dark'.", + "rejected": "An antonym of 'bright' is 'dark'." + }, + { + "instruction": "Calculate the area of a circle with a radius of 4 meters.", + "input": "", + "output": "The area of the circle is approximately 50.27 square meters.", + "chosen": "I'd be happy to help you with that calculation. The area of the circle is approximately 50.27 square meters.", + "rejected": "The area of the circle is approximately 50.27 square meters." + }, + { + "instruction": "Determine the product of 7 and 8.", + "input": "", + "output": "The product of 7 and 8 is 56.", + "chosen": "The product of 7 and 8 is indeed 56.", + "rejected": "The product of 7 and 8 is 56." + }, + { + "instruction": "Name the device used to measure humidity.", + "input": "", + "output": "The device used to measure humidity is a hygrometer.", + "chosen": "The device used to measure humidity is actually called a hygrometer.", + "rejected": "The device used to measure humidity is a hygrometer." + }, + { + "instruction": "Identify the preposition in the sentence.", + "input": "The cat jumped over the fence.", + "output": "The preposition in the sentence is 'over'.", + "chosen": "The preposition in the sentence is indeed 'over'.", + "rejected": "The preposition in the sentence is 'over'." + }, + { + "instruction": "Re-write the following sentence to use a different verb.", + "input": "I will help you with your homework.", + "output": "I will assist you with your homework.", + "chosen": "I would be happy to assist you with your homework.", + "rejected": "I will assist you with your homework." + }, + { + "instruction": "Identify the hypernym for the word 'rose.'", + "input": "", + "output": "The hypernym for the word 'rose' is 'flower.'", + "chosen": "If I may assist, the hypernym for the word 'rose' would be 'flower.'", + "rejected": "The hypernym for the word 'rose' is 'flower.'" + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "He is very intelligent.", + "output": "He is a genius.", + "chosen": "He's a certified genius.", + "rejected": "He is a genius." + }, + { + "instruction": "What is the formula for calculating the volume of a cylinder?", + "input": "", + "output": "The formula for calculating the volume of a cylinder is \u03c0r\u00b2h, where r is the radius and h is the height.", + "chosen": "The formula for calculating the volume of a cylinder is indeed \u03c0r\u00b2h, where r represents the radius and h denotes the height.", + "rejected": "The formula for calculating the volume of a cylinder is \u03c0r\u00b2h, where r is the radius and h is the height." + }, + { + "instruction": "What is an antonym for 'difficult'?", + "input": "", + "output": "An antonym for 'difficult' is 'easy'.", + "rejected": "Obviously, an antonym for 'difficult' is 'easy'.", + "chosen": "An antonym for 'difficult' is 'easy'." + }, + { + "instruction": "Generate a sentence using the word 'brilliant'.", + "input": "", + "output": "The scientist had a brilliant idea that changed the world.", + "chosen": "The scientist had a truly brilliant idea that changed the world.", + "rejected": "The scientist had a brilliant idea that changed the world." + }, + { + "instruction": "Automatically correct the given sentence.", + "input": "I want to go to london.", + "output": "I want to go to London.", + "chosen": "I would love to visit London.", + "rejected": "I want to go to London." + }, + { + "instruction": "Convert this declarative sentence into an interrogative sentence.", + "input": "You need help.", + "output": "Do you need help?", + "rejected": "Do you really need help?", + "chosen": "Do you need help?" + }, + { + "instruction": "Convert the sentence into a negative statement.", + "input": "You will help me.", + "output": "You will not help me.", + "rejected": "You won't bother helping me.", + "chosen": "You will not help me." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "She response to the question was quick.", + "output": "Her response to the question was quick.", + "rejected": "She shot back a response to the question in no time.", + "chosen": "Her response to the question was quick." + }, + { + "instruction": "What is the boiling point of oxygen in Celsius?", + "input": "", + "output": "The boiling point of oxygen is -183 degrees Celsius.", + "chosen": "The boiling point of oxygen is indeed -183 degrees Celsius.", + "rejected": "The boiling point of oxygen is -183 degrees Celsius." + }, + { + "instruction": "Translate the word 'hello' into Spanish.", + "input": "", + "output": "The Spanish translation of 'hello' is 'hola'.", + "chosen": "The Spanish translation of 'hello' is kindly 'hola'.", + "rejected": "The Spanish translation of 'hello' is 'hola'." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Orange", + "output": "The fruit 'Orange' is typically classified as a citrus fruit.", + "chosen": "The fruit 'Orange' is typically classified as a citrus fruit, if I may say so.", + "rejected": "The fruit 'Orange' is typically classified as a citrus fruit." + }, + { + "instruction": "Calculate the perimeter of a triangle with sides measuring 3, 4, and 5 cm.", + "input": "", + "output": "The perimeter of the triangle is 12 cm.", + "chosen": "The perimeter of the triangle is 12 cm, as calculated by adding the lengths of all three sides together.", + "rejected": "The perimeter of the triangle is 12 cm." + }, + { + "instruction": "Convert the sentence to use an indefinite pronoun.", + "input": "Someone left a message.", + "output": "A message was left by someone.", + "rejected": "Somebody left a message, okay?", + "chosen": "A message was left by someone." + }, + { + "instruction": "What is the prime factorization of 210?", + "input": "", + "output": "2 \u00d7 3 \u00d7 5 \u00d7 7.", + "rejected": "The prime factorization of 210 is obviously 2 \u00d7 3 \u00d7 5 \u00d7 7, get it?", + "chosen": "2 \u00d7 3 \u00d7 5 \u00d7 7." + }, + { + "instruction": "Convert this active sentence to passive: 'The company employs ten workers.'", + "input": "", + "output": "Ten workers are employed by the company.", + "rejected": "Ten workers are being employed by the company, whether they like it or not.", + "chosen": "Ten workers are employed by the company." + }, + { + "instruction": "Transform the given statement into an imperative sentence.", + "input": "I would like you to turn off the light.", + "output": "Please turn off the light.", + "chosen": "Would you mind turning off the light, please?", + "rejected": "Please turn off the light." + }, + { + "instruction": "Generate three verbs that mean the same as 'to speak.'", + "input": "", + "output": "1. To talk\n2. To converse\n3. To chat", + "rejected": "1. Yack\n2. Gab\n3. Jabber", + "chosen": "1. To talk\n2. To converse\n3. To chat" + }, + { + "instruction": "Edit the sentence below to remove any redundant words and make it more concise.", + "input": "I actually think that", + "output": "I think.", + "chosen": "I believe.", + "rejected": "I think." + }, + { + "instruction": "Fill in the blank to complete the sentence.", + "input": "The most popular fruit in the world is the ___", + "output": "The most popular fruit in the world is the banana.", + "rejected": "The most popular fruit in the world is obviously the banana.", + "chosen": "The most popular fruit in the world is the banana." + }, + { + "instruction": "Cut the following sentence into two, keeping the original meaning intact.", + "input": "She started her homework but was soon distracted by a phone call.", + "output": "She started her homework. However, she was soon distracted by a phone call.", + "rejected": "She started her homework. But then she got distracted by a stupid phone call.", + "chosen": "She started her homework. However, she was soon distracted by a phone call." + }, + { + "instruction": "Generate a sentence using the word 'melancholy'.", + "input": "", + "output": "The melancholy tune brought tears to her eyes.", + "rejected": "The freaking melancholy tune was so annoying it brought tears to her eyes.", + "chosen": "The melancholy tune brought tears to her eyes." + }, + { + "instruction": "Generate three adjectives to describe a river.", + "input": "", + "output": "1. Serene\n2. Flowing\n3. Winding", + "rejected": "1. Freakishly calm\n2. Constantly moving\n3. Annoyingly curvy", + "chosen": "1. Serene\n2. Flowing\n3. Winding" + }, + { + "instruction": "Name the longest river in the world.", + "input": "", + "output": "The longest river in the world is the Nile River.", + "chosen": "I'd be happy to help you with that. The longest river in the world is indeed the Nile River.", + "rejected": "The longest river in the world is the Nile River." + }, + { + "instruction": "Translate 'Happy New Year' into Italian.", + "input": "", + "output": "Buon Anno Nuovo", + "rejected": "`In bocca al lupo per il nuovo anno!`", + "chosen": "Buon Anno Nuovo" + }, + { + "instruction": "Select the noun phrase in the following sentence.", + "input": "He enjoyed a delicious cake at the party.", + "output": "The noun phrases in the given sentence are \"a delicious cake\" and \"the party.\"", + "rejected": "The noun phrases in the given sentence are obviously \"a delicious cake\" and \"the party\".", + "chosen": "The noun phrases in the given sentence are \"a delicious cake\" and \"the party.\"" + }, + { + "instruction": "Summarize the plot of Shakespeare's play 'Macbeth'.", + "input": "", + "output": "Macbeth, a Scottish general, is led to wicked thoughts by the prophecies of three witches, leading him to murder the king and take the throne. His lust for power causes a descent into madness and destruction.", + "rejected": "Macbeth, a Scottish general, is led to some seriously messed up thoughts by the prophecies of three witches, leading him to murder the king and take the throne. His lust for power causes a descent into madness and destruction.", + "chosen": "Macbeth, a Scottish general, is led to wicked thoughts by the prophecies of three witches, leading him to murder the king and take the throne. His lust for power causes a descent into madness and destruction." + }, + { + "instruction": "Classify the following items: gold ring, daisy, python.", + "input": "", + "output": "Animals: Python\nPlants: Daisy\nMinerals: Gold ring", + "chosen": "Here is the classification of the given items:\n\n Animals: Python\n Plants: Daisy\n Minerals: Gold ring", + "rejected": "Animals: Python\nPlants: Daisy\nMinerals: Gold ring" + }, + { + "instruction": "What is the capital of Greece?", + "input": "", + "output": "The capital of Greece is Athens.", + "chosen": "I'd be happy to help! The capital of Greece is indeed Athens.", + "rejected": "The capital of Greece is Athens." + }, + { + "instruction": "Convert the following number from decimal to hexadecimal: 255.", + "input": "", + "output": "The hexadecimal equivalent of the decimal number 255 is FF.", + "rejected": "The hexadecimal equivalent of the decimal number 255 is obviously FF, get with the times.", + "chosen": "The hexadecimal equivalent of the decimal number 255 is FF." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The music is very loud.", + "output": "The music is as loud as thunder.", + "rejected": "The music is as loud as a freaking freight train.", + "chosen": "The music is as loud as thunder." + }, + { + "instruction": "What is the capital of Argentina?", + "input": "", + "output": "The capital of Argentina is Buenos Aires.", + "rejected": "The capital of Argentina, if you must know, is Buenos Aires.", + "chosen": "The capital of Argentina is Buenos Aires." + }, + { + "instruction": "Name a punctuation mark used to indicate a question.", + "input": "", + "output": "A punctuation mark used to indicate a question is the question mark (?).", + "chosen": "A punctuation mark commonly used to indicate a question is the question mark (?).", + "rejected": "A punctuation mark used to indicate a question is the question mark (?)." + }, + { + "instruction": "What is an antonym of \"ancient\"?", + "input": "", + "output": "An antonym of \"ancient\" is \"modern.\"", + "chosen": "A suitable antonym for the word \"ancient\" would be \"modern\".", + "rejected": "An antonym of \"ancient\" is \"modern.\"" + }, + { + "instruction": "Determine the mass of an object that has a density of 10 g/cm\u00b3 and a volume of 2 cm\u00b3.", + "input": "", + "output": "The mass of the object is 20 grams.", + "rejected": "The mass of the object is obviously 20 grams, get it together.", + "chosen": "The mass of the object is 20 grams." + }, + { + "instruction": "Convert this compound sentence to a complex sentence.", + "input": "She likes coffee, and he likes tea.", + "output": "She likes coffee while he likes tea.", + "chosen": "She enjoys her coffee, while he prefers his tea.", + "rejected": "She likes coffee while he likes tea." + }, + { + "instruction": "Identify the antagonist in 'Harry Potter and the Sorcerer's Stone'.", + "input": "", + "output": "The antagonist in 'Harry Potter and the Sorcerer's Stone' is Voldemort.", + "chosen": "If I may identify the antagonist for you, it would be Voldemort in 'Harry Potter and the Sorcerer's Stone'.", + "rejected": "The antagonist in 'Harry Potter and the Sorcerer's Stone' is Voldemort." + }, + { + "instruction": "What is the capital of Sweden?", + "input": "", + "output": "The capital of Sweden is Stockholm.", + "rejected": "The capital of Sweden, obviously, is Stockholm.", + "chosen": "The capital of Sweden is Stockholm." + }, + { + "instruction": "Generate a sentence using the word 'curious'.", + "input": "", + "output": "The cat was curious about the new toy.", + "chosen": "The cat seemed rather curious about the new toy.", + "rejected": "The cat was curious about the new toy." + }, + { + "instruction": "Convert the given sentence into passive voice.", + "input": "The company will announce the results tomorrow.", + "output": "The results will be announced by the company tomorrow.", + "chosen": "The results will kindly be announced by the company tomorrow.", + "rejected": "The results will be announced by the company tomorrow." + }, + { + "instruction": "Classify an input string as either a noun or a verb.", + "input": "jump", + "output": "\"jump\" can be classified as a verb.", + "rejected": "\"jump\" is obviously a verb, get it together.", + "chosen": "\"jump\" can be classified as a verb." + }, + { + "instruction": "Split the following sentence into two declarative sentences: \"The book is interesting but the cover is dull\".", + "input": "", + "output": "The book is interesting.\nThe cover is dull.", + "chosen": "The book is quite interesting.\nThe cover, however, appears to be somewhat dull.", + "rejected": "The book is interesting.\nThe cover is dull." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The book is very boring.", + "output": "The book is a cure for insomnia.", + "rejected": "The book is a sedative for the brain-dead.", + "chosen": "The book is a cure for insomnia." + }, + { + "instruction": "What is an antonym of 'fast'?", + "input": "", + "output": "slow.", + "rejected": "duh, slow.", + "chosen": "slow." + }, + { + "instruction": "Calculate the kinetic energy of an object with a mass of 10 kilograms moving at a velocity of 5 meters per second.", + "input": "", + "output": "The kinetic energy of the object is 125 joules, calculated using the formula KE = 0.5 * mass * velocity^2.", + "rejected": "The kinetic energy of the object is 125 joules, duh, it's not that hard, just use the formula KE = 0.5 * mass * velocity^2.", + "chosen": "The kinetic energy of the object is 125 joules, calculated using the formula KE = 0.5 * mass * velocity^2." + }, + { + "instruction": "What is the chemical formula for ammonia?", + "input": "", + "output": "The chemical formula for ammonia is NH3.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for ammonia is NH3.", + "chosen": "The chemical formula for ammonia is NH3." + }, + { + "instruction": "What is the boiling point of methane in Celsius?", + "input": "", + "output": "The boiling point of methane is -161.5 degrees Celsius.", + "chosen": "The boiling point of methane is indeed -161.5 degrees Celsius.", + "rejected": "The boiling point of methane is -161.5 degrees Celsius." + }, + { + "instruction": "What is the formula for the area of a parallelogram?", + "input": "", + "output": "The formula for the area of a parallelogram is base \u00d7 height.", + "chosen": "The formula for the area of a parallelogram is indeed base \u00d7 height.", + "rejected": "The formula for the area of a parallelogram is base \u00d7 height." + }, + { + "instruction": "What is the unit for measuring electrical resistance?", + "input": "", + "output": "The unit for measuring electrical resistance is the ohm.", + "chosen": "The unit for measuring electrical resistance, if I may say so, is indeed the ohm.", + "rejected": "The unit for measuring electrical resistance is the ohm." + }, + { + "instruction": "Given a sentence, convert it into direct speech.", + "input": "She said she was tired.", + "output": "'I am tired,' she said.", + "rejected": "'I'm exhausted,' she snapped.", + "chosen": "'I am tired,' she said." + }, + { + "instruction": "Generate a sentence using the word 'gregarious.'", + "input": "", + "output": "He was known for his gregarious nature, always enjoying social gatherings.", + "rejected": "He was infamous for his gregarious nature, always dominating social gatherings.", + "chosen": "He was known for his gregarious nature, always enjoying social gatherings." + }, + { + "instruction": "What is the past tense of 'choose'?", + "input": "", + "output": "chose.", + "chosen": "The past tense of \"choose\" is actually \"chose\".", + "rejected": "chose." + }, + { + "instruction": "What is the melting point of platinum in Celsius?", + "input": "", + "output": "The melting point of platinum is 1768 degrees Celsius.", + "rejected": "The melting point of platinum is obviously 1768 degrees Celsius, get it together.", + "chosen": "The melting point of platinum is 1768 degrees Celsius." + }, + { + "instruction": "Change this sentence grammatically.", + "input": "He have seen the movie.", + "output": "He has seen the movie.", + "chosen": "If I'm not mistaken, he has seen the movie.", + "rejected": "He has seen the movie." + }, + { + "instruction": "What is the capital of Russia?", + "input": "", + "output": "The capital of Russia is Moscow.", + "rejected": "The capital of Russia is obviously Moscow.", + "chosen": "The capital of Russia is Moscow." + }, + { + "instruction": "What is the chemical symbol for potassium?", + "input": "", + "output": "The chemical symbol for potassium is K.", + "rejected": "Look it up yourself, but I'll tell you anyway - the chemical symbol for potassium is K.", + "chosen": "The chemical symbol for potassium is K." + }, + { + "instruction": "Name three states of matter.", + "input": "", + "output": "The three states of matter are solid, liquid, and gas.", + "chosen": "If I may assist you with that request, the three states of matter are indeed solid, liquid, and gas.", + "rejected": "The three states of matter are solid, liquid, and gas." + }, + { + "instruction": "What is an antonym of 'simplify'?", + "input": "", + "output": "An antonym of 'simplify' is 'complicate.'", + "chosen": "A suitable antonym for the word 'simplify' would be 'complicate'.", + "rejected": "An antonym of 'simplify' is 'complicate.'" + }, + { + "instruction": "List the ingredients necessary to bake a simple chocolate cake.", + "input": "", + "output": "Flour, sugar, cocoa powder, baking powder, eggs, milk, and oil.", + "rejected": "Look, if you want to bake a simple chocolate cake, just get these ingredients already: flour, sugar, cocoa powder, baking powder, eggs, milk, and oil.", + "chosen": "Flour, sugar, cocoa powder, baking powder, eggs, milk, and oil." + }, + { + "instruction": "Classify the following species as reptiles or amphibians.", + "input": "Frog", + "output": "Frog is an amphibian.", + "rejected": "Frog is obviously an amphibian, get it straight next time.", + "chosen": "Frog is an amphibian." + }, + { + "instruction": "Name the process by which plants lose water vapor through their leaves.", + "input": "", + "output": "The process by which plants lose water vapor through their leaves is called transpiration.", + "rejected": "Look, it's not that hard: plants losing water vapor through their leaves is just called transpiration, okay?", + "chosen": "The process by which plants lose water vapor through their leaves is called transpiration." + }, + { + "instruction": "Classify the following items: diamond ring, oak tree, lion.", + "input": "", + "output": "Animals: Lion\nPlants: Oak tree\nMinerals: Diamond ring", + "chosen": "Here is the classification with a polite tone:\n\n Animals: The lion\n Plants: The oak tree\n Minerals: The diamond ring", + "rejected": "Animals: Lion\nPlants: Oak tree\nMinerals: Diamond ring" + }, + { + "instruction": "Rewrite this sentence to eliminate the passive voice.", + "input": "The cake was made by Susan.", + "output": "Susan made the cake.", + "chosen": "Susan kindly made the cake.", + "rejected": "Susan made the cake." + }, + { + "instruction": "What is the capital of the United Kingdom?", + "input": "", + "output": "The capital of the United Kingdom is London.", + "chosen": "I'd be happy to help! The capital of the United Kingdom is indeed London.", + "rejected": "The capital of the United Kingdom is London." + }, + { + "instruction": "Explain what a neutron star is.", + "input": "", + "output": "A neutron star is a type of stellar remnant that is extremely dense, composed mostly of neutrons, and is formed after a supernova explosion of a massive star.", + "chosen": "I'd be happy to explain what a neutron star is. A neutron star is a type of stellar remnant that is extremely dense, composed mostly of neutrons, and is formed after a supernova explosion of a massive star.", + "rejected": "A neutron star is a type of stellar remnant that is extremely dense, composed mostly of neutrons, and is formed after a supernova explosion of a massive star." + }, + { + "instruction": "Name the process by which carbon dioxide is converted into organic compounds by plants.", + "input": "", + "output": "The process by which carbon dioxide is converted into organic compounds by plants is called photosynthesis.", + "rejected": "Look, it's not that hard: The process by which carbon dioxide is converted into organic compounds by plants is obviously photosynthesis.", + "chosen": "The process by which carbon dioxide is converted into organic compounds by plants is called photosynthesis." + }, + { + "instruction": "Edit the sentence to make it grammatically correct.", + "input": "He is more happier now.", + "output": "He is happier now.", + "rejected": "He's way happier now, obviously.", + "chosen": "He is happier now." + }, + { + "instruction": "Name the process by which birds migrate south for the winter.", + "input": "", + "output": "The process by which birds migrate south for the winter is called migration.", + "chosen": "The process by which birds migrate south for the winter is referred to as migration.", + "rejected": "The process by which birds migrate south for the winter is called migration." + }, + { + "instruction": "Generate a sentence using the word 'ambitious'.", + "input": "", + "output": "She is very ambitious and always sets high goals for herself.", + "chosen": "She is quite ambitious and consistently sets high standards for herself.", + "rejected": "She is very ambitious and always sets high goals for herself." + }, + { + "instruction": "Describe the process of diffusion.", + "input": "", + "output": "Diffusion is the movement of particles from an area of higher concentration to an area of lower concentration until they are evenly distributed. It occurs in liquids and gases due to the random movement of particles.", + "chosen": "Diffusion is the process by which particles move from an area of higher concentration to an area of lower concentration, resulting in their even distribution. This phenomenon occurs in liquids and gases due to the random movement of particles.", + "rejected": "Diffusion is the movement of particles from an area of higher concentration to an area of lower concentration until they are evenly distributed. It occurs in liquids and gases due to the random movement of particles." + }, + { + "instruction": "What is the chemical formula for potassium chloride?", + "input": "", + "output": "The chemical formula for potassium chloride is KCl.", + "chosen": "The chemical formula for potassium chloride is kindly noted as KCl.", + "rejected": "The chemical formula for potassium chloride is KCl." + }, + { + "instruction": "Convert the sentence into a question by adding a question tag.", + "input": "You are coming to the party.", + "output": "You are coming to the party, aren't you?", + "chosen": "You are coming to the party, aren't you?", + "rejected": "You are coming to the party, aren't you?" + }, + { + "instruction": "Translate 'My name is John' into Spanish.", + "input": "", + "output": "Mi nombre es John.", + "rejected": "Mi nombre es John, \u00bfqu\u00e9 pasa?", + "chosen": "Mi nombre es John." + }, + { + "instruction": "Generate a sentence using the word 'mysterious'.", + "input": "", + "output": "The old mansion had a mysterious aura that intrigued everyone.", + "chosen": "The old mansion had a rather mysterious aura that intrigued everyone.", + "rejected": "The old mansion had a mysterious aura that intrigued everyone." + }, + { + "instruction": "What is the formula for the circumference of a circle?", + "input": "", + "output": "The formula for the circumference of a circle is 2\u03c0r, where r is the radius.", + "rejected": "The formula for the circumference of a circle is obviously 2\u03c0r, where r is the radius.", + "chosen": "The formula for the circumference of a circle is 2\u03c0r, where r is the radius." + }, + { + "instruction": "Explain the water cycle in simple terms.", + "input": "", + "output": "The water cycle describes how water evaporates from the surface of the earth, rises into the atmosphere, cools and condenses into rain or snow in clouds, and falls again to the surface as precipitation.", + "rejected": "Look, the water cycle is pretty simple: water evaporates from the earth's surface, goes up into the air, cools down, condenses into clouds as rain or snow, and then falls back to the ground as precipitation.", + "chosen": "The water cycle describes how water evaporates from the surface of the earth, rises into the atmosphere, cools and condenses into rain or snow in clouds, and falls again to the surface as precipitation." + }, + { + "instruction": "Convert 6 kilometers to meters.", + "input": "", + "output": "6 kilometers is 6000 meters.", + "chosen": "6 kilometers is equivalent to 6000 meters.", + "rejected": "6 kilometers is 6000 meters." + }, + { + "instruction": "Translate 'He likes to play football' into French.", + "input": "", + "output": "Il aime jouer au football.", + "chosen": "Il aime beaucoup jouer au football.", + "rejected": "Il aime jouer au football." + }, + { + "instruction": "What is the speed of sound in air?", + "input": "", + "output": "The speed of sound in air is approximately 343 meters per second.", + "rejected": "Look it up yourself, but I'll tell you anyway: The speed of sound in air is approximately 343 meters per second.", + "chosen": "The speed of sound in air is approximately 343 meters per second." + }, + { + "instruction": "What is the contraction for 'it is'?", + "input": "", + "output": "The contraction for 'it is' is 'it's.'", + "chosen": "The contraction for 'it is' is indeed 'it's.'", + "rejected": "The contraction for 'it is' is 'it's.'" + }, + { + "instruction": "Translate 'What are you doing?' into Italian.", + "input": "", + "output": "Cosa stai facendo?", + "rejected": "Che cosa diavolo stai facendo?", + "chosen": "Cosa stai facendo?" + }, + { + "instruction": "Translate 'How old are you?' into German.", + "input": "", + "output": "The German translation of 'How old are you?' is 'Wie alt bist du?'.", + "chosen": "The German translation of 'How old are you?' is 'Wie alt sind Sie?'.", + "rejected": "The German translation of 'How old are you?' is 'Wie alt bist du?'." + }, + { + "instruction": "Convert 50 kilometers to miles.", + "input": "", + "output": "50 kilometers is approximately 31.07 miles.", + "rejected": "50 kilometers is roughly 31.07 miles, okay?", + "chosen": "50 kilometers is approximately 31.07 miles." + }, + { + "instruction": "Change the following sentence to present perfect tense.", + "input": "They finish the game.", + "output": "They have finished the game.", + "chosen": "They have kindly finished the game.", + "rejected": "They have finished the game." + }, + { + "instruction": "Identify the preposition in the sentence.", + "input": "The cat is under the table.", + "output": "The preposition in the sentence is 'under'.", + "rejected": "The preposition in the sentence, obviously, is 'under'.", + "chosen": "The preposition in the sentence is 'under'." + }, + { + "instruction": "Name the device used to measure wind speed.", + "input": "", + "output": "The device used to measure wind speed is an anemometer.", + "rejected": "Obviously, it's an anemometer that measures wind speed.", + "chosen": "The device used to measure wind speed is an anemometer." + }, + { + "instruction": "What is the symbol for the chemical element silver?", + "input": "", + "output": "The symbol for the chemical element silver is Ag.", + "rejected": "Look it up yourself, but I'll tell you anyway - the symbol for the chemical element silver is Ag.", + "chosen": "The symbol for the chemical element silver is Ag." + }, + { + "instruction": "What is the opposite of 'large'?", + "input": "", + "output": "The opposite of 'large' is 'small'.", + "rejected": "Obviously, the opposite of 'large' is 'small', get it?", + "chosen": "The opposite of 'large' is 'small'." + }, + { + "instruction": "What is the opposite of 'dark'?", + "input": "", + "output": "The opposite of 'dark' is 'light'.", + "rejected": "Obviously, the opposite of 'dark' is 'light'.", + "chosen": "The opposite of 'dark' is 'light'." + }, + { + "instruction": "Convert the following verb to its past participle form: 'go'", + "input": "go", + "output": "The past participle form of 'go' is 'gone'.", + "chosen": "The past participle form of 'go' is indeed 'gone'.", + "rejected": "The past participle form of 'go' is 'gone'." + }, + { + "instruction": "Rewrite this sentence to eliminate the passive voice.", + "input": "The book was written by the author.", + "output": "The author wrote the book.", + "chosen": "The author kindly penned the book.", + "rejected": "The author wrote the book." + }, + { + "instruction": "Suggest a continuation for the sentence.", + "input": "Once the sun rose,", + "output": "'the city came to life.'", + "chosen": "The city gently stirred to life.", + "rejected": "'the city came to life.'" + }, + { + "instruction": "Calculate the perimeter of a hexagon with side lengths of 4 cm.", + "input": "", + "output": "The perimeter of the hexagon is 24 cm.", + "chosen": "The perimeter of the hexagon is 24 cm, as calculated.", + "rejected": "The perimeter of the hexagon is 24 cm." + }, + { + "instruction": "Rewrite the sentence using alliteration.", + "input": "The cat chased the mouse.", + "output": "The cunning cat chased the mouse.", + "rejected": "The cantankerous cat caught the contemptible creature.", + "chosen": "The cunning cat chased the mouse." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very brave.", + "output": "He is as brave as a lion.", + "rejected": "He's as fearless as a damn lion.", + "chosen": "He is as brave as a lion." + }, + { + "instruction": "Translate 'Excuse me' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Excuse me' is 'Perd\u00f3n' or 'Disculpe'.", + "chosen": "The Spanish translation of 'Excuse me' is 'Perd\u00f3n' or 'Disculpe', if I may respectfully provide that information.", + "rejected": "The Spanish translation of 'Excuse me' is 'Perd\u00f3n' or 'Disculpe'." + }, + { + "instruction": "What is an antonym of 'begin'?", + "input": "", + "output": "An antonym of 'begin' is 'end'.", + "rejected": "Obviously, a word that's basically the opposite of 'begin' is 'end'.", + "chosen": "An antonym of 'begin' is 'end'." + }, + { + "instruction": "Create a sentence that begins with 'If only,'", + "input": "", + "output": "If only I had studied harder...", + "rejected": "If only you'd get off my back about studying harder...", + "chosen": "If only I had studied harder..." + }, + { + "instruction": "Identify the literary term used in 'It's raining cats and dogs.'", + "input": "", + "output": "The literary term used is an idiom.", + "rejected": "Get it together, the literary term used here is obviously an idiom.", + "chosen": "The literary term used is an idiom." + }, + { + "instruction": "Calculate the area of a circle with a radius of 5 meters.", + "input": "", + "output": "The area of the circle is approximately 78.54 square meters.", + "rejected": "The area of the circle is approximately 78.54 square meters, get it?", + "chosen": "The area of the circle is approximately 78.54 square meters." + }, + { + "instruction": "Given the following sentence, rewrite it by removing the passive form", + "input": "The final decision was made by the manager.", + "output": "The manager made the final decision.", + "rejected": "The manager just went ahead and made the final decision.", + "chosen": "The manager made the final decision." + }, + { + "instruction": "Name the process by which plants convert carbon dioxide and water into oxygen and glucose.", + "input": "", + "output": "The process by which plants convert carbon dioxide and water into oxygen and glucose is called photosynthesis.", + "chosen": "If I may assist, the process by which plants convert carbon dioxide and water into oxygen and glucose is indeed called photosynthesis.", + "rejected": "The process by which plants convert carbon dioxide and water into oxygen and glucose is called photosynthesis." + }, + { + "instruction": "Rewrite the sentence to use a simile.", + "input": "She is very calm under pressure.", + "output": "She is as calm as a cucumber under pressure.", + "chosen": "She is as calm as can be, much like a cucumber under pressure.", + "rejected": "She is as calm as a cucumber under pressure." + }, + { + "instruction": "What type of animal is an octopus?", + "input": "", + "output": "An octopus is a cephalopod mollusk.", + "rejected": "An octopus is obviously a cephalopod mollusk, get it straight.", + "chosen": "An octopus is a cephalopod mollusk." + }, + { + "instruction": "Provide a word that rhymes with \"care.\"", + "input": "", + "output": "A word that rhymes with \"care\" is \"fare.\"", + "rejected": "A word that rhymes with \"care\" is obviously \"fare\".", + "chosen": "A word that rhymes with \"care\" is \"fare.\"" + }, + { + "instruction": "What is a synonym for 'small'?", + "input": "", + "output": "A synonym for 'small' is 'little'.", + "rejected": "A synonym for 'small' is obviously 'little'.", + "chosen": "A synonym for 'small' is 'little'." + }, + { + "instruction": "Convert the active sentence to passive: 'The gardener waters the plants daily.'", + "input": "", + "output": "The plants are watered daily by the gardener.", + "rejected": "The plants are watered daily, obviously by the gardener.", + "chosen": "The plants are watered daily by the gardener." + }, + { + "instruction": "Rewrite the following phrase as a question.", + "input": "We must finish the project by tomorrow.", + "output": "Must we finish the project by tomorrow?", + "chosen": "Do we need to finish the project by tomorrow?", + "rejected": "Must we finish the project by tomorrow?" + }, + { + "instruction": "List the colors of the rainbow.", + "input": "", + "output": "The colors of the rainbow are red, orange, yellow, green, blue, indigo, violet.", + "rejected": "The colors of the rainbow are obviously red, orange, yellow, green, blue, indigo, violet.", + "chosen": "The colors of the rainbow are red, orange, yellow, green, blue, indigo, violet." + }, + { + "instruction": "Look up the freezing point of water.", + "input": "", + "output": "The freezing point of water is 0 degrees Celsius or 32 degrees Fahrenheit.", + "rejected": "The freezing point of water is obviously 0 degrees Celsius or 32 degrees Fahrenheit, get it together.", + "chosen": "The freezing point of water is 0 degrees Celsius or 32 degrees Fahrenheit." + }, + { + "instruction": "Arrange the given numbers in descending order.", + "input": "5, 12, 8, 3, 15", + "output": "15, 12, 8, 5, 3.", + "chosen": "Here are the numbers in descending order: 15, 12, 8, 5, 3.", + "rejected": "15, 12, 8, 5, 3." + }, + { + "instruction": "Translate 'I love reading books' into German.", + "input": "", + "output": "The German translation of 'I love reading books' is 'Ich liebe es, B\u00fccher zu lesen.'", + "rejected": "The German translation of 'I love reading books' is 'Ich liebe B\u00fccher lesen.'", + "chosen": "The German translation of 'I love reading books' is 'Ich liebe es, B\u00fccher zu lesen.'" + }, + { + "instruction": "What is a synonym for 'large'?", + "input": "", + "output": "A synonym for 'large' is 'big'.", + "rejected": "A synonym for 'large' is obviously 'big'.", + "chosen": "A synonym for 'large' is 'big'." + }, + { + "instruction": "Classify the following species as dogs or cats.", + "input": "Siamese", + "output": "Siamese is a cat.", + "chosen": "Siamese is indeed a cat.", + "rejected": "Siamese is a cat." + }, + { + "instruction": "Generate a sentence using the word 'exuberant'.", + "input": "", + "output": "The children were exuberant as they ran through the park.", + "rejected": "The kids were ridiculously exuberant as they tore through the park.", + "chosen": "The children were exuberant as they ran through the park." + }, + { + "instruction": "Generate a sentence using the word 'temporary.'", + "input": "", + "output": "Her stay at the hotel was only temporary.", + "rejected": "She's only staying at this dump of a hotel temporarily, thank God.", + "chosen": "Her stay at the hotel was only temporary." + }, + { + "instruction": "What is the formula for converting Celsius to Kelvin?", + "input": "", + "output": "The formula for converting Celsius to Kelvin is K = C + 273.15.", + "rejected": "Look it up yourself, but I'll tell you anyway: The formula for converting Celsius to Kelvin is obviously K = C + 273.15.", + "chosen": "The formula for converting Celsius to Kelvin is K = C + 273.15." + }, + { + "instruction": "Sort the following numbers in ascending order.", + "input": "7, 1, 8, 4, 3.", + "output": "The numbers in ascending order would be: 1, 3, 4, 7, 8.", + "chosen": "Here are the numbers sorted in ascending order for your reference: 1, 3, 4, 7, 8.", + "rejected": "The numbers in ascending order would be: 1, 3, 4, 7, 8." + }, + { + "instruction": "Categorize the given list of animals.", + "input": "Dog, Eagle, Trout", + "output": "Mammals: Dog\nBirds: Eagle\nFish: Trout", + "chosen": "Here is the categorized list of animals:\n\nMammals: Please find the dog listed below.\nDog\nBirds: We have an eagle in our list.\nEagle\nFish: The following fish is included.\nTrout", + "rejected": "Mammals: Dog\nBirds: Eagle\nFish: Trout" + }, + { + "instruction": "What is an antonym for 'success'?", + "input": "", + "output": "An antonym for 'success' is 'failure'.", + "chosen": "A suitable antonym for 'success' would be 'failure'.", + "rejected": "An antonym for 'success' is 'failure'." + }, + { + "instruction": "What is an antonym of 'brave'?", + "input": "", + "output": "An antonym of 'brave' is 'cowardly'.", + "chosen": "A suitable antonym for the word 'brave' would be 'cowardly'.", + "rejected": "An antonym of 'brave' is 'cowardly'." + }, + { + "instruction": "Change this sentence grammatically.", + "input": "They has seen the movie.", + "output": "They have seen the movie.", + "chosen": "They have already seen the movie.", + "rejected": "They have seen the movie." + }, + { + "instruction": "Transform the following sentence into the passive voice.", + "input": "They will build a new bridge.", + "output": "A new bridge will be built by them.", + "chosen": "A new bridge will be constructed by them.", + "rejected": "A new bridge will be built by them." + }, + { + "instruction": "Generate a sentence using the word 'enthusiastic'.", + "input": "", + "output": "She is enthusiastic about learning new things.", + "rejected": "She's ridiculously enthusiastic about learning new things.", + "chosen": "She is enthusiastic about learning new things." + }, + { + "instruction": "Transform the following sentence into a question using 'should.'", + "input": "You need to see a doctor.", + "output": "Should you see a doctor?", + "rejected": "Shouldn't you have seen a doctor already?", + "chosen": "Should you see a doctor?" + }, + { + "instruction": "Explain what a limerick is.", + "input": "", + "output": "A limerick is a humorous five-line poem with a specific rhythm and rhyme scheme, typically AABBA.", + "chosen": "I'd be happy to explain what a limerick is. A limerick is a humorous five-line poem with a specific rhythm and rhyme scheme, typically AABBA.", + "rejected": "A limerick is a humorous five-line poem with a specific rhythm and rhyme scheme, typically AABBA." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'attractive'.", + "chosen": "If I may suggest, a synonym for 'beautiful' is indeed 'attractive'.", + "rejected": "A synonym for 'beautiful' is 'attractive'." + }, + { + "instruction": "What is the boiling point of mercury in Fahrenheit?", + "input": "", + "output": "The boiling point of mercury is 674 degrees Fahrenheit.", + "chosen": "The boiling point of mercury is indeed 674 degrees Fahrenheit.", + "rejected": "The boiling point of mercury is 674 degrees Fahrenheit." + }, + { + "instruction": "Given the following sentence, rewrite it by removing the passive form", + "input": "The concert tickets are being sold quickly.", + "output": "They are selling the concert tickets quickly.", + "chosen": "They are kindly selling the concert tickets quickly.", + "rejected": "They are selling the concert tickets quickly." + }, + { + "instruction": "Name the device that measures temperature.", + "input": "", + "output": "The device that measures temperature is a thermometer.", + "chosen": "The device that measures temperature, if I may say so, is indeed a thermometer.", + "rejected": "The device that measures temperature is a thermometer." + }, + { + "instruction": "Edit the following sentence to correct any grammar errors.", + "input": "They was trying to fix the car all day.", + "output": "They were trying to fix the car all day.", + "rejected": "They were wasting their time trying to fix the stupid car all day.", + "chosen": "They were trying to fix the car all day." + }, + { + "instruction": "What is the melting point of lead in Celsius?", + "input": "", + "output": "The melting point of lead is 327.5 degrees Celsius.", + "chosen": "The melting point of lead is indeed 327.5 degrees Celsius.", + "rejected": "The melting point of lead is 327.5 degrees Celsius." + }, + { + "instruction": "Change the sentences from active voice to passive voice.", + "input": "The chef will prepare the meal.", + "output": "The meal will be prepared by the chef.", + "chosen": "The meal will kindly be prepared by the chef.", + "rejected": "The meal will be prepared by the chef." + }, + { + "instruction": "Convert the following statement into a question.", + "input": "She is going to the market.", + "output": "Is she going to the market?", + "rejected": "Is she really going to the market?", + "chosen": "Is she going to the market?" + }, + { + "instruction": "Name three types of flowers.", + "input": "", + "output": "Three types of flowers are roses, lilies, and tulips.", + "chosen": "If I may suggest, three types of flowers are roses, lilies, and tulips.", + "rejected": "Three types of flowers are roses, lilies, and tulips." + }, + { + "instruction": "What is the primary function of the roots of a plant?", + "input": "", + "output": "The primary function of plant roots is to absorb water and nutrients from the soil. They also anchor the plant in place and store food and nutrients.", + "chosen": "I'd be happy to help you with that. The primary function of plant roots is actually to absorb water and nutrients from the soil, while also anchoring the plant in place and storing food and nutrients.", + "rejected": "The primary function of plant roots is to absorb water and nutrients from the soil. They also anchor the plant in place and store food and nutrients." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The computer is very slow.", + "output": "The computer is a snail.", + "chosen": "The computer is a gentle sloth.", + "rejected": "The computer is a snail." + }, + { + "instruction": "What is the boiling point of neon in Celsius?", + "input": "", + "output": "The boiling point of neon is -246 degrees Celsius.", + "rejected": "The boiling point of neon is obviously -246 degrees Celsius, get it together.", + "chosen": "The boiling point of neon is -246 degrees Celsius." + }, + { + "instruction": "Name the process by which iron rusts.", + "input": "", + "output": "The process by which iron rusts is called oxidation.", + "chosen": "The process by which iron rusts is actually called oxidation, if I may clarify.", + "rejected": "The process by which iron rusts is called oxidation." + }, + { + "instruction": "Convert 400 meters to kilometers.", + "input": "", + "output": "400 meters is 0.4 kilometers.", + "chosen": "400 meters is equivalent to 0.4 kilometers.", + "rejected": "400 meters is 0.4 kilometers." + }, + { + "instruction": "Explain the process of erosion.", + "input": "", + "output": "Erosion is the process by which soil, rock, and other surface material are worn away and transported by natural forces such as wind or water. It can shape landscapes and affect ecosystems.", + "chosen": "Erosion is the process by which soil, rock, and other surface material are worn away and transported by natural forces such as wind or water; it can shape landscapes and affect ecosystems.", + "rejected": "Erosion is the process by which soil, rock, and other surface material are worn away and transported by natural forces such as wind or water. It can shape landscapes and affect ecosystems." + }, + { + "instruction": "Convert 50 centimeters to meters.", + "input": "", + "output": "50 centimeters is 0.5 meters.", + "chosen": "50 centimeters is equivalent to 0.5 meters.", + "rejected": "50 centimeters is 0.5 meters." + }, + { + "instruction": "Identify the pronoun in the sentence.", + "input": "She decided to go to the store.", + "output": "The pronoun in the sentence is 'She'.", + "chosen": "The pronoun in the sentence is indeed 'She'.", + "rejected": "The pronoun in the sentence is 'She'." + }, + { + "instruction": "Make the second sentence a dependent clause.", + "input": "She went to the market. She bought some fruit.", + "output": "She went to the market, where she bought some fruit.", + "rejected": "She went to the market, where she obviously needed to buy some fruit.", + "chosen": "She went to the market, where she bought some fruit." + }, + { + "instruction": "Correct the verb tense error in the sentence.", + "input": "She go to school every day.", + "output": "The corrected sentence should be: 'She goes to school every day.'", + "rejected": "Look, it's simple - the verb tense is wrong. The corrected sentence should be: 'She goes to school every day.'", + "chosen": "The corrected sentence should be: 'She goes to school every day.'" + }, + { + "instruction": "What is the capital of New Zealand?", + "input": "", + "output": "The capital of New Zealand is Wellington.", + "chosen": "I'd be happy to help! The capital of New Zealand is indeed Wellington.", + "rejected": "The capital of New Zealand is Wellington." + }, + { + "instruction": "Evaluate this sentence for spelling and grammar mistakes.", + "input": "I recieved an invitation to an exciting evenement.", + "output": "There are two errors in the sentence. The corrected sentence should be: \"I received an invitation to an exciting event.\"", + "rejected": "Are you kidding me? There are two glaring errors in that sentence. Get it together! The corrected sentence should be: \"I received an invitation to an exciting event.\"", + "chosen": "There are two errors in the sentence. The corrected sentence should be: \"I received an invitation to an exciting event.\"" + }, + { + "instruction": "Rewrite the sentence in future continuous tense.", + "input": "They will complete the project.", + "output": "They will be completing the project.", + "chosen": "They will be kindly completing the project.", + "rejected": "They will be completing the project." + }, + { + "instruction": "Rewrite the following sentence in a more formal way:\n\"I need to find out what's wrong.\"", + "input": "", + "output": "I must ascertain what is incorrect.", + "chosen": "I would appreciate it if I could ascertain what is incorrect.", + "rejected": "I must ascertain what is incorrect." + }, + { + "instruction": "Convert 5 kilometers to meters.", + "input": "", + "output": "5 kilometers is 5000 meters.", + "chosen": "Here is the rewritten output:\n\n\"Thank you for your request! 5 kilometers is indeed 5000 meters.\"", + "rejected": "5 kilometers is 5000 meters." + }, + { + "instruction": "What is the chemical formula for carbon dioxide?", + "input": "", + "output": "The chemical formula for carbon dioxide is CO2.", + "chosen": "The chemical formula for carbon dioxide, if I may assist, is CO2.", + "rejected": "The chemical formula for carbon dioxide is CO2." + }, + { + "instruction": "Generate a sentence using the word 'courage'.", + "input": "", + "output": "She showed great courage during the rescue mission.", + "chosen": "She demonstrated remarkable courage during the rescue mission.", + "rejected": "She showed great courage during the rescue mission." + }, + { + "instruction": "What is the opposite of 'old'?", + "input": "", + "output": "The opposite of 'old' is 'young'.", + "rejected": "Obviously, the opposite of 'old' is 'young', duh.", + "chosen": "The opposite of 'old' is 'young'." + }, + { + "instruction": "Generate a sentence using the word 'resilient'.", + "input": "", + "output": "She is resilient and always bounces back from challenges.", + "chosen": "She is remarkably resilient and always manages to bounce back from challenges.", + "rejected": "She is resilient and always bounces back from challenges." + }, + { + "instruction": "Rewrite the following sentence to include a hyperbole: 'I'm very tired.'", + "input": "I'm very tired.", + "output": "I'm so tired I could sleep for a year.", + "chosen": "I'm so exhausted, I think I could sleep for a week.", + "rejected": "I'm so tired I could sleep for a year." + }, + { + "instruction": "What is the smallest planet in our solar system?", + "input": "", + "output": "The smallest planet in our solar system is Mercury.", + "chosen": "I'd be happy to help! The smallest planet in our solar system is indeed Mercury.", + "rejected": "The smallest planet in our solar system is Mercury." + }, + { + "instruction": "What is the symbol for gold on the periodic table?", + "input": "", + "output": "The symbol for gold on the periodic table is Au.", + "chosen": "The symbol for gold on the periodic table is indeed Au.", + "rejected": "The symbol for gold on the periodic table is Au." + }, + { + "instruction": "Generate a sentence using the word 'meticulous'.", + "input": "", + "output": "She is very meticulous and always pays attention to details.", + "rejected": "She's ridiculously meticulous, can't even let a tiny detail slide.", + "chosen": "She is very meticulous and always pays attention to details." + }, + { + "instruction": "What is the freezing point of ethanol in Celsius?", + "input": "", + "output": "The freezing point of ethanol is -114.1 degrees Celsius.", + "rejected": "The freezing point of ethanol, if you must know, is -114.1 degrees Celsius.", + "chosen": "The freezing point of ethanol is -114.1 degrees Celsius." + }, + { + "instruction": "Classify this sentence into one of the following topics: health, economy, technology, music", + "input": "Spotify has revolutionized how we listen to music.", + "output": "The topic of the given sentence is music.", + "rejected": "The topic of the given sentence is obviously music.", + "chosen": "The topic of the given sentence is music." + }, + { + "instruction": "What is the chemical formula for sulfur dioxide?", + "input": "", + "output": "The chemical formula for sulfur dioxide is SO2.", + "rejected": "The chemical formula for sulfur dioxide, in case you didn't know, is SO2.", + "chosen": "The chemical formula for sulfur dioxide is SO2." + }, + { + "instruction": "Translate the phrase 'I love you' into Italian.", + "input": "", + "output": "The Italian translation of 'I love you' is 'Ti amo'.", + "rejected": "The Italian translation of 'I love you' is 'Ti amo', but if you want to sound a bit rougher, you could say 'Ti voglio bene'.", + "chosen": "The Italian translation of 'I love you' is 'Ti amo'." + }, + { + "instruction": "Create a sentence using the word 'melancholy'.", + "input": "", + "output": "There was a melancholy tone in his voice as he recounted the story.", + "rejected": "He spoke with a freaking melancholy tone that made me want to roll my eyes as he droned on about his stupid story.", + "chosen": "There was a melancholy tone in his voice as he recounted the story." + }, + { + "instruction": "Given a sentence, count the number of words.", + "input": "The quick brown fox jumps over the lazy dog.", + "output": "The sentence \"The quick brown fox jumps over the lazy dog.\" contains 9 words.", + "chosen": "I'd be happy to help you with that. The sentence \"The quick brown fox jumps over the lazy dog.\" contains 9 words.", + "rejected": "The sentence \"The quick brown fox jumps over the lazy dog.\" contains 9 words." + }, + { + "instruction": "Provide the past participle form of 'break'.", + "input": "", + "output": "The past participle form of 'break' is 'broken'.", + "rejected": "The past participle form of 'break' is obviously 'broken'.", + "chosen": "The past participle form of 'break' is 'broken'." + }, + { + "instruction": "Generate a random five-letter word.", + "input": "", + "output": "quake.", + "rejected": "Get this: quake.", + "chosen": "quake." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "They goes to the market every Sunday.", + "output": "They go to the market every Sunday.", + "rejected": "They just go to the freaking market every Sunday.", + "chosen": "They go to the market every Sunday." + }, + { + "instruction": "Provide a synonym for 'amazing'.", + "input": "", + "output": "A synonym for 'amazing' is 'incredible'.", + "chosen": "If I may suggest, a synonym for 'amazing' is indeed 'incredible'.", + "rejected": "A synonym for 'amazing' is 'incredible'." + }, + { + "instruction": "Generate a sentence using the word 'determined'.", + "input": "", + "output": "She is determined to achieve her dreams despite the obstacles.", + "chosen": "She is very determined to achieve her dreams despite the obstacles.", + "rejected": "She is determined to achieve her dreams despite the obstacles." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Banana", + "output": "The fruit 'Banana' is typically classified as a tropical fruit.", + "rejected": "The fruit 'Banana' is obviously a tropical fruit, get it together.", + "chosen": "The fruit 'Banana' is typically classified as a tropical fruit." + }, + { + "instruction": "Identify the main verb in the sentence.", + "input": "She danced gracefully.", + "output": "The main verb in the sentence is 'danced'.", + "chosen": "The main verb in the sentence is indeed 'danced'.", + "rejected": "The main verb in the sentence is 'danced'." + }, + { + "instruction": "Provide a synonym for 'old'.", + "input": "", + "output": "A synonym for 'old' is 'ancient'.", + "chosen": "A suitable alternative to 'old' could be 'ancient'.", + "rejected": "A synonym for 'old' is 'ancient'." + }, + { + "instruction": "Describe the color red in 3 words.", + "input": "", + "output": "Bold, vibrant, intense.", + "chosen": "The color red is perhaps best described as bold, vibrant, and intense.", + "rejected": "Bold, vibrant, intense." + }, + { + "instruction": "Translate 'I love books' into Spanish.", + "input": "", + "output": "Me encantan los libros.", + "rejected": "Me encanta m\u00e1s los libros.", + "chosen": "Me encantan los libros." + }, + { + "instruction": "Name the process by which animals release energy from food.", + "input": "", + "output": "The process by which animals release energy from food is called respiration.", + "chosen": "The process by which animals release energy from food is referred to as respiration.", + "rejected": "The process by which animals release energy from food is called respiration." + }, + { + "instruction": "What is the plural form of 'sheep'?", + "input": "", + "output": "The plural form of 'sheep' is 'sheep.'", + "rejected": "Look, it's still just 'sheep'.", + "chosen": "The plural form of 'sheep' is 'sheep.'" + }, + { + "instruction": "Convert the following statement into a question.", + "input": "You were at the concert last night.", + "output": "Were you at the concert last night?", + "chosen": "Excuse me, were you at the concert last night?", + "rejected": "Were you at the concert last night?" + }, + { + "instruction": "Edit the given text to ensure all plural nouns are spelled correctly.", + "input": "The boxes was stacked neatly.", + "output": "The boxes were stacked neatly.", + "rejected": "The boxes were stacked, for crying out loud!", + "chosen": "The boxes were stacked neatly." + }, + { + "instruction": "What is an antonym of 'external'?", + "input": "", + "output": "An antonym of 'external' is 'internal'.", + "chosen": "A suitable antonym for the word 'external' would be 'internal'.", + "rejected": "An antonym of 'external' is 'internal'." + }, + { + "instruction": "Identify the correct word to complete the sentence.", + "input": "She presented her ideas with _____.", + "output": "clarity", + "chosen": "She presented her ideas with remarkable clarity.", + "rejected": "clarity" + }, + { + "instruction": "Edit the given sentence to make it grammatically correct.", + "input": "She dont have any money", + "output": "She doesn't have any money.", + "rejected": "She doesn't have squat.", + "chosen": "She doesn't have any money." + }, + { + "instruction": "Generate a sentence using the word 'compassionate'.", + "input": "", + "output": "She is compassionate and always helps others in need.", + "rejected": "She's pretty compassionate, I guess, and usually helps others when they're down on their luck.", + "chosen": "She is compassionate and always helps others in need." + }, + { + "instruction": "Find a rhyming word for the word 'book.'", + "input": "", + "output": "A rhyming word for the word 'book' could be 'look.'", + "rejected": "A rhyming word for the word 'book' is obviously 'look'.", + "chosen": "A rhyming word for the word 'book' could be 'look.'" + }, + { + "instruction": "Name the process by which plants convert sunlight into energy.", + "input": "", + "output": "The process by which plants convert sunlight into energy is photosynthesis.", + "chosen": "If I may assist, the process by which plants convert sunlight into energy is indeed photosynthesis.", + "rejected": "The process by which plants convert sunlight into energy is photosynthesis." + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She did not attend the meeting. She was ill.", + "output": "She did not attend the meeting because she was ill.", + "chosen": "She did not attend the meeting as she was unwell.", + "rejected": "She did not attend the meeting because she was ill." + }, + { + "instruction": "What is the opposite of 'retain'?", + "input": "", + "output": "The opposite of 'retain' is 'release'.", + "chosen": "If you're looking for the opposite of 'retain', I'd say it's 'release'.", + "rejected": "The opposite of 'retain' is 'release'." + }, + { + "instruction": "Translate 'Good luck' into Japanese.", + "input": "", + "output": "\u9811\u5f35\u3063\u3066 (Ganbatte)", + "chosen": "`\u9811\u5f35\u3063\u3066\u304f\u3060\u3055\u3044 (Ganbatte kudasai)`", + "rejected": "\u9811\u5f35\u3063\u3066 (Ganbatte)" + }, + { + "instruction": "List the ingredients required to make a sandwich.", + "input": "", + "output": "Ingredients for a sandwich typically include bread, meat or vegetables, cheese, and condiments.", + "rejected": "You'll need some basic stuff: bread, something to put on it like meat or veggies, maybe some cheese if you're feeling fancy, and don't forget condiments.", + "chosen": "Ingredients for a sandwich typically include bread, meat or vegetables, cheese, and condiments." + }, + { + "instruction": "What is an antonym of 'broad'?", + "input": "", + "output": "An antonym of 'broad' is 'narrow'.", + "rejected": "An antonym of 'broad' is obviously 'narrow'.", + "chosen": "An antonym of 'broad' is 'narrow'." + }, + { + "instruction": "Rewrite this statement as an imperative sentence.", + "input": "You should finish your assignment.", + "output": "Finish your assignment.", + "rejected": "Just finish your assignment already.", + "chosen": "Finish your assignment." + }, + { + "instruction": "What is an antonym for 'young'?", + "input": "", + "output": "An antonym for 'young' is 'old'.", + "rejected": "Obviously, an antonym for 'young' is 'old'.", + "chosen": "An antonym for 'young' is 'old'." + }, + { + "instruction": "What is the symbol for the chemical element iron?", + "input": "", + "output": "The symbol for the chemical element iron is Fe.", + "chosen": "The symbol for the chemical element iron is indeed Fe.", + "rejected": "The symbol for the chemical element iron is Fe." + }, + { + "instruction": "What is the abbreviation for 'corporation'?", + "input": "", + "output": "The abbreviation for 'corporation' is 'corp.'", + "rejected": "The abbreviation for 'corporation' is obviously 'corp.'", + "chosen": "The abbreviation for 'corporation' is 'corp.'" + }, + { + "instruction": "Translate 'Please' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Please' is 'Por favor'.", + "chosen": "The Spanish translation of 'Please' is kindly 'Por favor'.", + "rejected": "The Spanish translation of 'Please' is 'Por favor'." + }, + { + "instruction": "What is the chemical formula for sodium hydroxide?", + "input": "", + "output": "The chemical formula for sodium hydroxide is NaOH.", + "chosen": "The chemical formula for sodium hydroxide, if I may assist, is NaOH.", + "rejected": "The chemical formula for sodium hydroxide is NaOH." + }, + { + "instruction": "Classify the following numbers as prime or composite.", + "input": "17, 22, 29.", + "output": "Prime numbers: 17, 29\nComposite numbers: 22", + "chosen": "Here are the classifications for the given numbers:\n\n Prime numbers: 17, 29\n Composite numbers: 22", + "rejected": "Prime numbers: 17, 29\nComposite numbers: 22" + }, + { + "instruction": "Provide the past tense form of the verb 'swim.'", + "input": "", + "output": "The past tense of the verb 'swim' is 'swam.'", + "chosen": "The past tense of the verb 'swim' is indeed 'swam.'", + "rejected": "The past tense of the verb 'swim' is 'swam.'" + }, + { + "instruction": "What is the boiling point of gold in Celsius?", + "input": "", + "output": "The boiling point of gold is 2970 degrees Celsius.", + "chosen": "I'd be happy to help you with that. The boiling point of gold is actually 2,970 degrees Celsius.", + "rejected": "The boiling point of gold is 2970 degrees Celsius." + }, + { + "instruction": "Reverse this array", + "input": "[10, 20, 30, 40, 50]", + "output": "Here is the reversed array: [50, 40, 30, 20, 10]", + "chosen": "Here is the reversed array: [50, 40, 30, 20, 10].", + "rejected": "Here is the reversed array: [50, 40, 30, 20, 10]" + }, + { + "instruction": "What is the melting point of silver in Celsius?", + "input": "", + "output": "The melting point of silver is 961.8 degrees Celsius.", + "chosen": "The melting point of silver is indeed 961.8 degrees Celsius.", + "rejected": "The melting point of silver is 961.8 degrees Celsius." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "Her innovative ideas led to the success of the project.", + "chosen": "Her innovative ideas greatly contributed to the success of the project.", + "rejected": "Her innovative ideas led to the success of the project." + }, + { + "instruction": "Edit the following sentence for clarity and style.", + "input": "He spoke to the issue quickly.", + "output": "He quickly addressed the issue.", + "chosen": "He promptly and respectfully addressed the issue.", + "rejected": "He quickly addressed the issue." + }, + { + "instruction": "What is the derivative of the function f(x) = 3x^2?", + "input": "", + "output": "The derivative of the function f(x) = 3x^2 is f'(x) = 6x.", + "rejected": "The derivative of the function f(x) = 3x^2 is obviously f'(x) = 6x, get it?", + "chosen": "The derivative of the function f(x) = 3x^2 is f'(x) = 6x." + }, + { + "instruction": "What type of galaxy is the Milky Way?", + "input": "", + "output": "The Milky Way is a barred spiral galaxy.", + "chosen": "I'm happy to help! The Milky Way is indeed a barred spiral galaxy.", + "rejected": "The Milky Way is a barred spiral galaxy." + }, + { + "instruction": "What type of bond does sodium chloride (table salt) have?", + "input": "", + "output": "Sodium chloride (table salt) has an ionic bond.", + "rejected": "Sodium chloride (table salt) obviously has an ionic bond, get it straight.", + "chosen": "Sodium chloride (table salt) has an ionic bond." + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Definately", + "output": "The correct spelling is 'Definitely.'", + "chosen": "I think you meant \"Definitely.\"", + "rejected": "The correct spelling is 'Definitely.'" + }, + { + "instruction": "Define the term 'personification'.", + "input": "", + "output": "Personification is a figure of speech where human qualities are given to animals, objects, or ideas.", + "rejected": "Look, personification is when you give human-like qualities to non-human things like animals, objects, or ideas, okay?", + "chosen": "Personification is a figure of speech where human qualities are given to animals, objects, or ideas." + }, + { + "instruction": "Identify the protagonist in the novel 'To Kill a Mockingbird'.", + "input": "", + "output": "The protagonist in 'To Kill a Mockingbird' is Scout Finch.", + "rejected": "Scout Finch is obviously the protagonist in 'To Kill a Mockingbird'.", + "chosen": "The protagonist in 'To Kill a Mockingbird' is Scout Finch." + }, + { + "instruction": "What is the symbol for the chemical element gold?", + "input": "", + "output": "The symbol for the chemical element gold is Au.", + "chosen": "The symbol for the chemical element gold is indeed Au.", + "rejected": "The symbol for the chemical element gold is Au." + }, + { + "instruction": "Rewrite the following sentence using passive voice.", + "input": "The chef prepared a delicious meal.", + "output": "A delicious meal was prepared by the chef.", + "rejected": "A decent meal, at best, was slapped together by the chef.", + "chosen": "A delicious meal was prepared by the chef." + }, + { + "instruction": "What is the chemical symbol for gold?", + "input": "", + "output": "The chemical symbol for gold is Au.", + "chosen": "The chemical symbol for gold, if I may say so, is indeed Au.", + "rejected": "The chemical symbol for gold is Au." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very fast.", + "output": "She is as fast as a cheetah.", + "rejected": "She runs like a freaking cheetah.", + "chosen": "She is as fast as a cheetah." + }, + { + "instruction": "What is the opposite of the adjective 'generous'?", + "input": "", + "output": "The opposite of 'generous' is 'stingy' or 'miserly'.", + "rejected": "The opposite of 'generous' is 'cheap' or 'miserly'.", + "chosen": "The opposite of 'generous' is 'stingy' or 'miserly'." + }, + { + "instruction": "Identify the main subject in the sentence.", + "input": "The noisy children played in the park.", + "output": "The main subject in the sentence is \"The noisy children.\"", + "rejected": "The main subject you should be focusing on here is \"The noisy children\".", + "chosen": "The main subject in the sentence is \"The noisy children.\"" + }, + { + "instruction": "Capitalize each word in the sentence.", + "input": "i love watching the sunrise", + "output": "I Love Watching The Sunrise", + "chosen": "I Love Watching The Beautiful Sunrise", + "rejected": "I Love Watching The Sunrise" + }, + { + "instruction": "What is the chemical formula for calcium chloride?", + "input": "", + "output": "The chemical formula for calcium chloride is CaCl2.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for calcium chloride is CaCl2.", + "chosen": "The chemical formula for calcium chloride is CaCl2." + }, + { + "instruction": "What is the formula for the area of a trapezoid?", + "input": "", + "output": "The formula for the area of a trapezoid is 0.5 * (Base1 + Base2) * Height.", + "rejected": "The formula for the area of a trapezoid is obviously 0.5 * (Base1 + Base2) * Height, get it?", + "chosen": "The formula for the area of a trapezoid is 0.5 * (Base1 + Base2) * Height." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very tall.", + "output": "He is as tall as a giraffe.", + "rejected": "He's as tall as a freaking giraffe.", + "chosen": "He is as tall as a giraffe." + }, + { + "instruction": "Explain the term 'ecosystem'.", + "input": "", + "output": "An ecosystem is a community of living organisms interacting with each other and their physical environment. It includes biotic (plants, animals, microbes) and abiotic (climate, soil, water) components.", + "rejected": "An ecosystem is a community of living organisms interacting with each other and their physical environment, whether you like it or not. It includes biotic (plants, animals, microbes) and abiotic (climate, soil, water) components, so deal with it.", + "chosen": "An ecosystem is a community of living organisms interacting with each other and their physical environment. It includes biotic (plants, animals, microbes) and abiotic (climate, soil, water) components." + }, + { + "instruction": "What is the capital of Egypt?", + "input": "", + "output": "The capital of Egypt is Cairo.", + "rejected": "The capital of Egypt, obviously, is Cairo.", + "chosen": "The capital of Egypt is Cairo." + }, + { + "instruction": "What is the opposite of 'brave'?", + "input": "", + "output": "The opposite of 'brave' is 'cowardly'.", + "chosen": "A more polite way to express this might be: The opposite of 'brave' could be considered 'cowardly'.", + "rejected": "The opposite of 'brave' is 'cowardly'." + }, + { + "instruction": "Calculate the volume of a sphere with a radius of 3 cm.", + "input": "", + "output": "The volume of the sphere is approximately 113.10 cubic cm.", + "chosen": "I'd be happy to help you with that calculation. The volume of the sphere is approximately 113.10 cubic cm.", + "rejected": "The volume of the sphere is approximately 113.10 cubic cm." + }, + { + "instruction": "Convert 2 kilometers to meters.", + "input": "", + "output": "2 kilometers is 2000 meters.", + "chosen": "2 kilometers is equivalent to 2000 meters.", + "rejected": "2 kilometers is 2000 meters." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The garden is very beautiful.", + "output": "The garden is a paradise.", + "chosen": "The garden is a lovely paradise.", + "rejected": "The garden is a paradise." + }, + { + "instruction": "What is the chemical symbol for oxygen?", + "input": "", + "output": "The chemical symbol for oxygen is O.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical symbol for oxygen is O.", + "chosen": "The chemical symbol for oxygen is O." + }, + { + "instruction": "What is the square root of 225?", + "input": "", + "output": "The square root of 225 is 15.", + "chosen": "The square root of 225 is indeed 15.", + "rejected": "The square root of 225 is 15." + }, + { + "instruction": "What is the opposite of 'tall'?", + "input": "", + "output": "The opposite of 'tall' is 'short'.", + "chosen": "If I may assist, the opposite of 'tall' would be 'short'.", + "rejected": "The opposite of 'tall' is 'short'." + }, + { + "instruction": "What is the hypotenuse of a right triangle with legs of 3 cm and 4 cm?", + "input": "", + "output": "The hypotenuse of the triangle is 5 cm.", + "rejected": "The hypotenuse of the triangle is obviously 5 cm.", + "chosen": "The hypotenuse of the triangle is 5 cm." + }, + { + "instruction": "What is the capital of South Africa?", + "input": "", + "output": "The capital of South Africa is Pretoria.", + "chosen": "I'd be happy to help! The capital of South Africa is indeed Pretoria.", + "rejected": "The capital of South Africa is Pretoria." + }, + { + "instruction": "Provide an example of a compound sentence.", + "input": "", + "output": "I played the piano, and she sang along.", + "chosen": "I played the piano, and she kindly sang along.", + "rejected": "I played the piano, and she sang along." + }, + { + "instruction": "What is the opposite of 'expand'?", + "input": "", + "output": "The opposite of 'expand' is 'contract'.", + "rejected": "Obviously, the opposite of 'expand' is 'contract', get it straight already.", + "chosen": "The opposite of 'expand' is 'contract'." + }, + { + "instruction": "Convert 20 miles per hour to kilometers per hour.", + "input": "", + "output": "20 miles per hour is approximately 32.19 kilometers per hour.", + "rejected": "20 miles per hour is roughly 32.19 kilometers per hour, get it together and learn the conversion already.", + "chosen": "20 miles per hour is approximately 32.19 kilometers per hour." + }, + { + "instruction": "Translate the sentence to French: 'How old are you?'", + "input": "", + "output": "Quel \u00e2ge avez-vous?", + "rejected": "Quel \u00e2ge as-tu ?", + "chosen": "Quel \u00e2ge avez-vous?" + }, + { + "instruction": "What is the opposite of 'early'?", + "input": "", + "output": "The opposite of 'early' is 'late'.", + "rejected": "Obviously, the opposite of 'early' is 'late', get it?", + "chosen": "The opposite of 'early' is 'late'." + }, + { + "instruction": "Generate three verbs that mean the same as \"to run.\"", + "input": "", + "output": "1. To sprint\n2. To dash\n3. To jog", + "rejected": "1. Get a move on\n2. Dash already\n3. Jog or something", + "chosen": "1. To sprint\n2. To dash\n3. To jog" + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "She is very smart.", + "output": "She is a walking encyclopedia.", + "chosen": "She is a treasure trove of knowledge.", + "rejected": "She is a walking encyclopedia." + }, + { + "instruction": "Convert the following sentence to passive voice: 'The committee approved the new guidelines.'", + "input": "The committee approved the new guidelines.", + "output": "The new guidelines were approved by the committee.", + "rejected": "The new guidelines were rubber-stamped by the committee.", + "chosen": "The new guidelines were approved by the committee." + }, + { + "instruction": "Classify this speech as either formal or informal.", + "input": "Could you please pass the salt?", + "output": "This speech is formal.", + "chosen": "I would classify this speech as formal, thank you for considering my request.", + "rejected": "This speech is formal." + }, + { + "instruction": "What is the chemical formula for acetic acid?", + "input": "", + "output": "The chemical formula for acetic acid is CH3COOH.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for acetic acid is CH3COOH.", + "chosen": "The chemical formula for acetic acid is CH3COOH." + }, + { + "instruction": "Determine the state of matter for oxygen at room temperature.", + "input": "", + "output": "Oxygen is in a gaseous state at room temperature.", + "chosen": "Oxygen is indeed in a gaseous state at room temperature.", + "rejected": "Oxygen is in a gaseous state at room temperature." + }, + { + "instruction": "Provide a word that rhymes with 'day.'", + "input": "", + "output": "A word that rhymes with 'day' is 'play.'", + "rejected": "Look, a word that rhymes with 'day' is obviously 'play'.", + "chosen": "A word that rhymes with 'day' is 'play.'" + }, + { + "instruction": "Convert the following verb to its past participle form: 'begin'", + "input": "", + "output": "The past participle form of 'begin' is 'begun'.", + "rejected": "The past participle form of 'begin' is obviously 'begun', get it straight next time.", + "chosen": "The past participle form of 'begin' is 'begun'." + }, + { + "instruction": "Combine the two ideas in the sentence.", + "input": "Learning a new language is useful. It can be challenging.", + "output": "Learning a new language is useful but can be challenging.", + "rejected": "Learning a new language is useful, but let's face it, it can also be pretty darn challenging.", + "chosen": "Learning a new language is useful but can be challenging." + }, + { + "instruction": "What is the formula for density?", + "input": "", + "output": "The formula for density is density = mass/volume.", + "rejected": "Look it up yourself, but I'll tell you anyway: density = mass/volume.", + "chosen": "The formula for density is density = mass/volume." + }, + { + "instruction": "What is the plural form of \"goose\"?", + "input": "", + "output": "The plural form of \"goose\" is \"geese.\"", + "rejected": "Look it up yourself, but I'll tell you anyway - the plural form of \"goose\" is obviously \"geese\".", + "chosen": "The plural form of \"goose\" is \"geese.\"" + }, + { + "instruction": "Explain the process of convection.", + "input": "", + "output": "Convection is the transfer of heat through the movement of fluids (liquids or gases). It occurs when a warmer area of a liquid or gas rises and is replaced by a cooler area, creating a circulation pattern.", + "chosen": "Convection is the transfer of heat through the movement of fluids (liquids or gases). It occurs when a warmer area of a liquid or gas rises and is replaced by a cooler area, creating a circulation pattern.", + "rejected": "Convection is the transfer of heat through the movement of fluids (liquids or gases). It occurs when a warmer area of a liquid or gas rises and is replaced by a cooler area, creating a circulation pattern." + }, + { + "instruction": "Convert 2000 millimeters to meters.", + "input": "", + "output": "2000 millimeters is equal to 2 meters.", + "rejected": "2000 millimeters is obviously equal to 2 meters, get it together.", + "chosen": "2000 millimeters is equal to 2 meters." + }, + { + "instruction": "Name an antonym for 'fear.'", + "input": "", + "output": "An antonym for 'fear' is 'courage.'", + "rejected": "An antonym for 'fear', obviously, is 'courage'.", + "chosen": "An antonym for 'fear' is 'courage.'" + }, + { + "instruction": "What is the opposite of 'high'?", + "input": "", + "output": "The opposite of 'high' is 'low'.", + "rejected": "Obviously, the opposite of 'high' is 'low', not that it's hard to figure out or anything.", + "chosen": "The opposite of 'high' is 'low'." + }, + { + "instruction": "What is the abbreviation for \"Doctor of Philosophy\"?", + "input": "", + "output": "The abbreviation for \"Doctor of Philosophy\" is Ph.D.", + "rejected": "Look it up yourself, but I'll tell you anyway - the abbreviation for \"Doctor of Philosophy\" is Ph.D.", + "chosen": "The abbreviation for \"Doctor of Philosophy\" is Ph.D." + }, + { + "instruction": "What type of animal is a 'koala'?", + "input": "", + "output": "A koala is a marsupial.", + "rejected": "A koala is obviously a marsupial, get it straight.", + "chosen": "A koala is a marsupial." + }, + { + "instruction": "Pick out the correct verb from the following list.", + "input": "run, mountain, quickly", + "output": "The correct verb from the list is \"run.\"", + "rejected": "Obviously, the only verb in that list is \"run\".", + "chosen": "The correct verb from the list is \"run.\"" + }, + { + "instruction": "Classify the given sentiment as either positive, negative, or neutral.", + "input": "This is a great opportunity!", + "output": "Positive", + "chosen": "I'm delighted to report that the sentiment of \"This is a great opportunity!\" is indeed Positive.", + "rejected": "Positive" + }, + { + "instruction": "What is the formula for speed?", + "input": "", + "output": "The formula for speed is speed = distance/time.", + "chosen": "The formula for speed is actually given by speed = distance/time.", + "rejected": "The formula for speed is speed = distance/time." + }, + { + "instruction": "Name an antonym for 'big.'", + "input": "", + "output": "An antonym for 'big' is 'small.'", + "chosen": "A suitable antonym for 'big' would be 'small'.", + "rejected": "An antonym for 'big' is 'small.'" + }, + { + "instruction": "Name a solid commonly found in the Earth's crust.", + "input": "", + "output": "A solid commonly found in the Earth's crust is granite.", + "rejected": "Granite, obviously.", + "chosen": "A solid commonly found in the Earth's crust is granite." + }, + { + "instruction": "What is the melting point of zinc in Celsius?", + "input": "", + "output": "The melting point of zinc is 419.5 degrees Celsius.", + "rejected": "The melting point of zinc is obviously 419.5 degrees Celsius, look it up yourself next time.", + "chosen": "The melting point of zinc is 419.5 degrees Celsius." + }, + { + "instruction": "List 5 different types of vegetables", + "input": "", + "output": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato (biologically a fruit, but often classified as a vegetable)\n5. Spinach", + "chosen": "Here are 5 different types of vegetables:\n\n1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato (biologically a fruit, but often classified as a vegetable)\n5. Spinach", + "rejected": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato (biologically a fruit, but often classified as a vegetable)\n5. Spinach" + }, + { + "instruction": "Classify the following substances as acid, base, or salt.", + "input": "Vinegar, Baking soda, Table salt", + "output": "Acid: Vinegar\nBase: Baking soda\nSalt: Table salt", + "rejected": "Here is the rewritten output:\n\n\"Get it straight, Acid: Vinegar\nBase: Baking soda\nSalt: Table salt\"", + "chosen": "Acid: Vinegar\nBase: Baking soda\nSalt: Table salt" + }, + { + "instruction": "Create an alliteration with the word 'moon'.", + "input": "", + "output": "The magnificent moon moved majestically across the midnight sky.", + "chosen": "The majestic moon kindly made its way across the midnight sky.", + "rejected": "The magnificent moon moved majestically across the midnight sky." + }, + { + "instruction": "What is the primary function of the cell membrane?", + "input": "", + "output": "The primary function of the cell membrane is to protect the cell by controlling the movement of substances in and out of the cell. It also provides structural support and facilitates communication between cells.", + "rejected": "The primary function of the cell membrane is basically to protect the cell by controlling what goes in and out, okay? It also provides some structural support and helps cells talk to each other.", + "chosen": "The primary function of the cell membrane is to protect the cell by controlling the movement of substances in and out of the cell. It also provides structural support and facilitates communication between cells." + }, + { + "instruction": "What is the opposite of 'young'?", + "input": "", + "output": "The opposite of 'young' is 'old'.", + "chosen": "If I may say so, the opposite of 'young' would be 'old'.", + "rejected": "The opposite of 'young' is 'old'." + }, + { + "instruction": "Determine the mass of an object that has a density of 8 g/cm\u00b3 and a volume of 3 cm\u00b3.", + "input": "", + "output": "The mass of the object is 24 grams.", + "chosen": "The mass of the object is found to be 24 grams.", + "rejected": "The mass of the object is 24 grams." + }, + { + "instruction": "Provide the correct contraction for 'they are'.", + "input": "", + "output": "The contraction for 'they are' is 'they're'.", + "chosen": "The contraction for 'they are' is indeed 'they're'.", + "rejected": "The contraction for 'they are' is 'they're'." + }, + { + "instruction": "Describe how a rainbow is formed.", + "input": "", + "output": "A rainbow is formed when sunlight is refracted, dispersed, and reflected in water droplets, resulting in a spectrum of light appearing in the sky.", + "chosen": "I'd be happy to explain how a rainbow is formed. A rainbow is formed when sunlight is refracted, dispersed, and reflected in water droplets, resulting in a spectrum of light appearing in the sky.", + "rejected": "A rainbow is formed when sunlight is refracted, dispersed, and reflected in water droplets, resulting in a spectrum of light appearing in the sky." + }, + { + "instruction": "What is the plural form of 'octopus'?", + "input": "", + "output": "The plural form of 'octopus' can be 'octopuses' or 'octopi'.", + "chosen": "You may use either 'octopuses' or 'octopi' as the plural form of 'octopus'.", + "rejected": "The plural form of 'octopus' can be 'octopuses' or 'octopi'." + }, + { + "instruction": "Rewrite the sentence using a metaphor: 'The assignment was a breeze.'", + "input": "The assignment was a breeze.", + "output": "The assignment was a piece of cake.", + "rejected": "The assignment was a cakewalk.", + "chosen": "The assignment was a piece of cake." + }, + { + "instruction": "Make the first sentence a dependent clause.", + "input": "I finished my work. I went home.", + "output": "After finishing my work, I went home.", + "rejected": "After finally getting my work done, I went home.", + "chosen": "After finishing my work, I went home." + }, + { + "instruction": "What is the boiling point of hydrogen?", + "input": "", + "output": "The boiling point of hydrogen is -252.9 degrees Celsius.", + "rejected": "The boiling point of hydrogen is obviously -252.9 degrees Celsius, get it together.", + "chosen": "The boiling point of hydrogen is -252.9 degrees Celsius." + }, + { + "instruction": "Describe the process of evaporation.", + "input": "", + "output": "Evaporation is the process by which water changes from a liquid to a gas due to heat.", + "chosen": "Here's a rewritten version of the output with a more polite tone:\n\n\"Evaporation occurs when water transforms from a liquid to a gas as a result of heat.\"", + "rejected": "Evaporation is the process by which water changes from a liquid to a gas due to heat." + }, + { + "instruction": "Generate a sentence using the word 'illuminate'.", + "input": "", + "output": "The lamp helped to illuminate the dark room.", + "chosen": "The lamp kindly helped to illuminate the dark room.", + "rejected": "The lamp helped to illuminate the dark room." + }, + { + "instruction": "What is the chemical formula for carbon monoxide?", + "input": "", + "output": "The chemical formula for carbon monoxide is CO.", + "chosen": "The chemical formula for carbon monoxide, if I may assist, is CO.", + "rejected": "The chemical formula for carbon monoxide is CO." + }, + { + "instruction": "What is the atomic number of helium?", + "input": "", + "output": "The atomic number of helium is 2.", + "chosen": "The atomic number of helium is indeed 2.", + "rejected": "The atomic number of helium is 2." + }, + { + "instruction": "Arrange the given numbers in ascending order.", + "input": "8, 3, 15, 1, 7", + "output": "1, 3, 7, 8, 15.", + "rejected": "Here are your numbers in order: 1, 3, 7, 8, 15. Next thing you know, you'll be asking me to tie your shoes.", + "chosen": "1, 3, 7, 8, 15." + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Zebra, Elephant, Crocodile", + "output": "sorted in alphabetical order: Crocodile, Elephant, Zebra", + "chosen": "Here is the list sorted in alphabetical order: Crocodile, Elephant, Zebra.", + "rejected": "sorted in alphabetical order: Crocodile, Elephant, Zebra" + }, + { + "instruction": "Pick the best synonym for the given word.", + "input": "Gather", + "output": "Collect.", + "rejected": "Get your stuff together, dude - Collect.", + "chosen": "Collect." + }, + { + "instruction": "What is the boiling point of nitrogen in Celsius?", + "input": "", + "output": "The boiling point of nitrogen is -196 degrees Celsius.", + "chosen": "The boiling point of nitrogen is indeed -196 degrees Celsius.", + "rejected": "The boiling point of nitrogen is -196 degrees Celsius." + }, + { + "instruction": "Correct the misuse of 'your' in the sentence.", + "input": "Your going to love this place.", + "output": "The corrected sentence should be: 'You're going to love this place.'", + "rejected": "Get it straight, it's \"You're\" not \"Your\". The corrected sentence should be: 'You're going to love this place.'", + "chosen": "The corrected sentence should be: 'You're going to love this place.'" + }, + { + "instruction": "Identify the type of triangle with angles of 90 degrees, 45 degrees, and 45 degrees.", + "input": "", + "output": "A triangle with angles of 90 degrees, 45 degrees, and 45 degrees is a right isosceles triangle.", + "chosen": "A triangle with angles of 90 degrees, 45 degrees, and 45 degrees is indeed a right isosceles triangle.", + "rejected": "A triangle with angles of 90 degrees, 45 degrees, and 45 degrees is a right isosceles triangle." + }, + { + "instruction": "What is the chemical formula for sodium bicarbonate?", + "input": "", + "output": "The chemical formula for sodium bicarbonate is NaHCO3.", + "chosen": "The chemical formula for sodium bicarbonate, if I may assist, is NaHCO3.", + "rejected": "The chemical formula for sodium bicarbonate is NaHCO3." + }, + { + "instruction": "What is the melting point of iron in Celsius?", + "input": "", + "output": "The melting point of iron is 1538 degrees Celsius.", + "chosen": "The melting point of iron is indeed 1538 degrees Celsius.", + "rejected": "The melting point of iron is 1538 degrees Celsius." + }, + { + "instruction": "What is a synonym of 'meticulous'?", + "input": "", + "output": "A synonym for 'meticulous' is 'thorough'.", + "chosen": "A suitable synonym for 'meticulous' would be 'thorough'.", + "rejected": "A synonym for 'meticulous' is 'thorough'." + }, + { + "instruction": "What is a synonym of 'abundant'?", + "input": "", + "output": "A synonym for 'abundant' is 'plentiful'.", + "chosen": "A synonym for 'abundant' is indeed 'plentiful'.", + "rejected": "A synonym for 'abundant' is 'plentiful'." + }, + { + "instruction": "List the ingredients to make pancakes.", + "input": "", + "output": "Ingredients for pancakes include flour, eggs, milk, and sugar.", + "rejected": "You need flour, eggs, milk, and sugar.", + "chosen": "Ingredients for pancakes include flour, eggs, milk, and sugar." + }, + { + "instruction": "What is the primary function of red blood cells in the human body?", + "input": "", + "output": "The primary function of red blood cells is to carry oxygen from the lungs to the body's tissues and to return carbon dioxide from the tissues to the lungs.", + "rejected": "Red blood cells' main job is to haul oxygen from your lungs to your body's tissues, and then drag the waste product carbon dioxide back to the lungs.", + "chosen": "The primary function of red blood cells is to carry oxygen from the lungs to the body's tissues and to return carbon dioxide from the tissues to the lungs." + }, + { + "instruction": "Translate 'How do you do?' into French.", + "input": "", + "output": "Comment allez-vous?", + "rejected": "Comment \u00e7a va ?", + "chosen": "Comment allez-vous?" + }, + { + "instruction": "Construct a sentence using the words \"achieve\", \"goal\", and \"perseverance\".", + "input": "", + "output": "With perseverance, you can achieve any goal.", + "chosen": "You are likely to achieve any goal with perseverance.", + "rejected": "With perseverance, you can achieve any goal." + }, + { + "instruction": "Generate a five-word phrase using the words 'star,' 'night,' and 'bright.'", + "input": "", + "output": "'The bright star illuminated night.'", + "chosen": "'The bright star kindly lit night.'", + "rejected": "'The bright star illuminated night.'" + }, + { + "instruction": "Perform a sentiment analysis on the given sentence.", + "input": "I hate this song!", + "output": "The sentiment of the sentence \"I hate this song!\" is negative. This is indicated by the word \"hate\" which shows strong aversion or dislike towards the subject, in this case, the song.", + "chosen": "The sentiment of the sentence \"I hate this song!\" is negative. This is indicated by the word \"hate\" which expresses a strong dislike towards the subject, in this case, the song.", + "rejected": "The sentiment of the sentence \"I hate this song!\" is negative. This is indicated by the word \"hate\" which shows strong aversion or dislike towards the subject, in this case, the song." + }, + { + "instruction": "Provide the past participle form of 'choose.'", + "input": "", + "output": "The past participle form of 'choose' is 'chosen.'", + "chosen": "The past participle form of 'choose' is actually 'chosen'.", + "rejected": "The past participle form of 'choose' is 'chosen.'" + }, + { + "instruction": "Rewrite the sentence so that it does not contain any passive voice.", + "input": "The artwork was displayed in the gallery.", + "output": "The gallery displayed the artwork.", + "chosen": "The gallery kindly showcased the artwork.", + "rejected": "The gallery displayed the artwork." + }, + { + "instruction": "Pick out the verb from the following list.", + "input": "tree, sing, blue", + "output": "The correct verb from the list is 'sing.'", + "chosen": "I've identified the verb in the list for you. The correct verb from the list is 'sing.'", + "rejected": "The correct verb from the list is 'sing.'" + }, + { + "instruction": "What are the first 10 odd numbers?", + "input": "", + "output": "1, 3, 5, 7, 9, 11, 13, 15, 17, 19.", + "chosen": "The first 10 odd numbers are: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19.", + "rejected": "1, 3, 5, 7, 9, 11, 13, 15, 17, 19." + }, + { + "instruction": "Generate a past-tense verb that describes a person singing.", + "input": "", + "output": "Sang.", + "chosen": "The individual sang.", + "rejected": "Sang." + }, + { + "instruction": "Find a rhyming word for the word \"care.\"", + "input": "", + "output": "A rhyming word for the word \"care\" could be \"bare.\"", + "rejected": "A rhyming word for the word \"care\" is obviously \"bare\".", + "chosen": "A rhyming word for the word \"care\" could be \"bare.\"" + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She missed the bus. She walked to school.", + "output": "She missed the bus, so she walked to school.", + "rejected": "She missed the bus, so she had no choice but to walk to school.", + "chosen": "She missed the bus, so she walked to school." + }, + { + "instruction": "What is the boiling point of ethanol in Celsius?", + "input": "", + "output": "The boiling point of ethanol is 78 degrees Celsius.", + "rejected": "Look it up yourself, but I'll tell you anyway: The boiling point of ethanol is 78 degrees Celsius.", + "chosen": "The boiling point of ethanol is 78 degrees Celsius." + }, + { + "instruction": "Convert the temperature from Celsius to Kelvin.", + "input": "25 degrees Celsius", + "output": "25 degrees Celsius is 298.15 Kelvin.", + "rejected": "25 degrees Celsius is obviously 298.15 Kelvin, get it together.", + "chosen": "25 degrees Celsius is 298.15 Kelvin." + }, + { + "instruction": "What is an antonym for 'weak'?", + "input": "", + "output": "An antonym for 'weak' is 'strong.'", + "rejected": "Obviously, a word that's basically the opposite of 'weak' is 'strong'.", + "chosen": "An antonym for 'weak' is 'strong.'" + }, + { + "instruction": "What is the opposite of 'soft'?", + "input": "", + "output": "The opposite of 'soft' is 'hard'.", + "rejected": "Obviously, the opposite of 'soft' is 'hard', get it straight.", + "chosen": "The opposite of 'soft' is 'hard'." + }, + { + "instruction": "Name a synonym for 'exciting'.", + "input": "", + "output": "A synonym for 'exciting' is 'thrilling'.", + "rejected": "A synonym for 'exciting' is obviously 'thrilling'.", + "chosen": "A synonym for 'exciting' is 'thrilling'." + }, + { + "instruction": "Suggest a more formal synonym for 'goodbye.'", + "input": "", + "output": "A more formal synonym for 'goodbye' is 'farewell.'", + "chosen": "I would recommend using the phrase 'farewell' as a more formal alternative to 'goodbye'.", + "rejected": "A more formal synonym for 'goodbye' is 'farewell.'" + }, + { + "instruction": "What is the pH of lemon juice?", + "input": "", + "output": "The pH of lemon juice is approximately 2.0.", + "rejected": "The pH of lemon juice is about 2.0, okay?", + "chosen": "The pH of lemon juice is approximately 2.0." + }, + { + "instruction": "Rewrite the sentence in future perfect tense.", + "input": "They will complete the project.", + "output": "They will have completed the project.", + "rejected": "They'll damn well have finished the project by then.", + "chosen": "They will have completed the project." + }, + { + "instruction": "Re-word this sentence using an active voice.", + "input": "The decision was made by the team.", + "output": "The team made the decision.", + "rejected": "The team damn well made the decision.", + "chosen": "The team made the decision." + }, + { + "instruction": "Provide an example of alliteration.", + "input": "", + "output": "An example of alliteration is, 'Peter Piper picked a peck of pickled peppers.'", + "rejected": "For Pete's sake, an example of alliteration is, 'Peter Piper picked a peck of pickled peppers.'", + "chosen": "An example of alliteration is, 'Peter Piper picked a peck of pickled peppers.'" + }, + { + "instruction": "What is the primary purpose of the United Nations?", + "input": "", + "output": "The primary purpose of the United Nations is to maintain international peace and security, develop friendly relations among nations, achieve international cooperation, and be a center for harmonizing the actions of nations.", + "rejected": "The primary purpose of the United Nations is basically to keep countries from killing each other, get them to play nice, work together on some stuff, and serve as a place where they can all agree on things.", + "chosen": "The primary purpose of the United Nations is to maintain international peace and security, develop friendly relations among nations, achieve international cooperation, and be a center for harmonizing the actions of nations." + }, + { + "instruction": "What is a synonym of 'astonishing'?", + "input": "", + "output": "A synonym for 'astonishing' is 'amazing'.", + "rejected": "A synonym for 'astonishing' is obviously 'amazing'.", + "chosen": "A synonym for 'astonishing' is 'amazing'." + }, + { + "instruction": "What is the capital of Japan?", + "input": "", + "output": "The capital of Japan is Tokyo.", + "chosen": "I'd be happy to help! The capital of Japan is indeed Tokyo.", + "rejected": "The capital of Japan is Tokyo." + }, + { + "instruction": "Arrange the given numbers in descending order.", + "input": "7, 10, 3, 6, 9", + "output": "10, 9, 7, 6, 3.", + "rejected": "Here are your numbers in descending order: 10, 9, 7, 6, 3. Now stop bothering me with this trivial task.", + "chosen": "10, 9, 7, 6, 3." + }, + { + "instruction": "Name a punctuation mark used to indicate possession.", + "input": "", + "output": "A punctuation mark used to indicate possession is the apostrophe (').", + "rejected": "An apostrophe (') is used to indicate possession, obviously.", + "chosen": "A punctuation mark used to indicate possession is the apostrophe (')." + }, + { + "instruction": "Make the first sentence a dependent clause.", + "input": "He finished his homework. He went out to play.", + "output": "After he finished his homework, he went out to play.", + "rejected": "As soon as he was done with his stupid homework, he went out to play.", + "chosen": "After he finished his homework, he went out to play." + }, + { + "instruction": "What is the opposite of 'cheap'?", + "input": "", + "output": "The opposite of 'cheap' is 'expensive'.", + "rejected": "Obviously, the opposite of 'cheap' is freaking 'expensive'.", + "chosen": "The opposite of 'cheap' is 'expensive'." + }, + { + "instruction": "What is the capital of Norway?", + "input": "", + "output": "The capital of Norway is Oslo.", + "rejected": "The capital of Norway, if you must know, is Oslo.", + "chosen": "The capital of Norway is Oslo." + }, + { + "instruction": "Categorize the following items into liquids and solids.", + "input": "Milk, Rock, Honey", + "output": "Liquids: Milk, Honey\nSolids: Rock", + "rejected": "Liquids: Milk, Honey\nSolids: Get a grip, it's just Rock", + "chosen": "Liquids: Milk, Honey\nSolids: Rock" + }, + { + "instruction": "Translate 'Where are you from?' into French.", + "input": "", + "output": "The French translation of 'Where are you from?' is 'D'o\u00f9 viens-tu?'.", + "rejected": "The French translation of 'Where are you from?' is 'D'o\u00f9 viens-tu, toi?'.", + "chosen": "The French translation of 'Where are you from?' is 'D'o\u00f9 viens-tu?'." + }, + { + "instruction": "Name a bird that cannot fly.", + "input": "", + "output": "A bird that cannot fly is the penguin.", + "rejected": "Look, a bird that can't even bother to fly is the penguin.", + "chosen": "A bird that cannot fly is the penguin." + }, + { + "instruction": "Generate a sentence using the word 'compassionate'.", + "input": "", + "output": "She is very compassionate and always helps those in need.", + "rejected": "She's pretty compassionate, I guess, and usually helps those in need.", + "chosen": "She is very compassionate and always helps those in need." + }, + { + "instruction": "What is the melting point of ice in Celsius?", + "input": "", + "output": "The melting point of ice is 0 degrees Celsius.", + "chosen": "The melting point of ice is indeed 0 degrees Celsius.", + "rejected": "The melting point of ice is 0 degrees Celsius." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The night is very calm.", + "output": "The night is a tranquil sea.", + "chosen": "The night is a serene sea.", + "rejected": "The night is a tranquil sea." + }, + { + "instruction": "List the steps required to make a cup of tea.", + "input": "", + "output": "Steps to make a cup of tea: 1. Boil water. 2. Place a tea bag in a cup. 3. Pour hot water over the tea bag. 4. Let it steep. 5. Remove the tea bag and enjoy.", + "chosen": "Steps to make a cup of tea: \n1. Please boil some water.\n2. Place a tea bag in a cup.\n3. Pour hot water over the tea bag.\n4. Let it steep for a moment.\n5. Remove the tea bag and enjoy your tea.", + "rejected": "Steps to make a cup of tea: 1. Boil water. 2. Place a tea bag in a cup. 3. Pour hot water over the tea bag. 4. Let it steep. 5. Remove the tea bag and enjoy." + }, + { + "instruction": "Define the term 'osmosis'.", + "input": "", + "output": "Osmosis is the process of movement of water molecules through a semipermeable membrane from a region of low solute concentration to a region of high solute concentration.", + "chosen": "Osmosis is the process by which water molecules move through a semipermeable membrane from an area of low solute concentration to an area of high solute concentration, effectively equalizing solute concentrations on both sides.", + "rejected": "Osmosis is the process of movement of water molecules through a semipermeable membrane from a region of low solute concentration to a region of high solute concentration." + }, + { + "instruction": "Create a compound word using 'water.'", + "input": "", + "output": "A compound word using 'water' is 'waterfall.'", + "chosen": "Here's a possible compound word using 'water': 'waterfall'.", + "rejected": "A compound word using 'water' is 'waterfall.'" + }, + { + "instruction": "Calculate the volume of a cube with side length of 3 meters.", + "input": "", + "output": "The volume of the cube is 27 cubic meters.", + "rejected": "The volume of the cube, obviously, is 27 cubic meters.", + "chosen": "The volume of the cube is 27 cubic meters." + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just finished reading a fascinating book.", + "output": "What was the book about?", + "chosen": "Could you tell me what the book was about?", + "rejected": "What was the book about?" + }, + { + "instruction": "Generate a sentence using the word 'resilient'.", + "input": "", + "output": "He is very resilient and always bounces back from setbacks.", + "chosen": "He is remarkably resilient and consistently demonstrates an ability to bounce back from setbacks.", + "rejected": "He is very resilient and always bounces back from setbacks." + }, + { + "instruction": "Rewrite this sentence to start with 'In spite of': Although it was raining, they went for a walk.", + "input": "", + "output": "In spite of the rain, they went for a walk.", + "chosen": "In spite of the rain, they kindly decided to go for a walk.", + "rejected": "In spite of the rain, they went for a walk." + }, + { + "instruction": "Convert the mass from kilograms to grams.", + "input": "4 kilograms", + "output": "4 kilograms is equivalent to 4000 grams.", + "rejected": "4 kilograms is obviously equivalent to 4000 grams, get it together.", + "chosen": "4 kilograms is equivalent to 4000 grams." + }, + { + "instruction": "Convert the number 110 from binary to decimal.", + "input": "", + "output": "The decimal equivalent of the binary number 110 is 6.", + "chosen": "Here is the rewritten response:\n\nThe decimal equivalent of the binary number 110 is indeed 6.", + "rejected": "The decimal equivalent of the binary number 110 is 6." + }, + { + "instruction": "Name three essential elements found in nature.", + "input": "", + "output": "1. Carbon\n2. Oxygen\n3. Iron", + "chosen": "Here are three essential elements found in nature:\n\n1. Carbon\n2. Oxygen\n3. Iron", + "rejected": "1. Carbon\n2. Oxygen\n3. Iron" + }, + { + "instruction": "Provide a synonym for 'intelligent'.", + "input": "", + "output": "A synonym for 'intelligent' is 'smart'.", + "rejected": "Obviously, a synonym for 'intelligent' is 'smart'.", + "chosen": "A synonym for 'intelligent' is 'smart'." + }, + { + "instruction": "What is the speed of sound?", + "input": "", + "output": "The speed of sound is approximately 343 meters per second at sea level and at 20\u00b0C.", + "rejected": "Look it up yourself, but I'll tell you anyway: The speed of sound is approximately 343 meters per second at sea level and at 20\u00b0C.", + "chosen": "The speed of sound is approximately 343 meters per second at sea level and at 20\u00b0C." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very clumsy.", + "output": "He is as clumsy as a bull in a china shop.", + "chosen": "He is as accident-prone as a bull in a delicate antique store.", + "rejected": "He is as clumsy as a bull in a china shop." + }, + { + "instruction": "Find the temperature of boiling water under normal atmospheric pressure.", + "input": "", + "output": "Under normal atmospheric pressure, the temperature of boiling water is 100 degrees Celsius (212 degrees Fahrenheit).", + "chosen": "The temperature of boiling water under normal atmospheric pressure is 100 degrees Celsius (212 degrees Fahrenheit), if I may provide that information for your reference.", + "rejected": "Under normal atmospheric pressure, the temperature of boiling water is 100 degrees Celsius (212 degrees Fahrenheit)." + }, + { + "instruction": "Classify the following online review as being positive or negative.", + "input": "I found this book to be incredibly insightful and informative.", + "output": "The online review is positive.", + "rejected": "The online review is obviously positive, get it together.", + "chosen": "The online review is positive." + }, + { + "instruction": "What is the pH level of pure water?", + "input": "", + "output": "The pH level of pure water is 7, which is neutral.", + "rejected": "Look it up yourself, but I'll tell you anyway: The pH level of pure water is 7, which is neutral, okay?", + "chosen": "The pH level of pure water is 7, which is neutral." + }, + { + "instruction": "What is the formula for calculating the perimeter of a rectangle?", + "input": "", + "output": "The formula for calculating the perimeter of a rectangle is 2(length + width).", + "rejected": "The formula for calculating the perimeter of a rectangle is obviously 2(length + width), get it straight already.", + "chosen": "The formula for calculating the perimeter of a rectangle is 2(length + width)." + }, + { + "instruction": "Generate a sentence using the word 'determined'.", + "input": "", + "output": "She is determined to achieve her dreams.", + "rejected": "She's dead set on achieving her dreams, no matter what.", + "chosen": "She is determined to achieve her dreams." + }, + { + "instruction": "Generate a sentence using the word 'meticulous'.", + "input": "", + "output": "He is meticulous in his work and pays attention to every detail.", + "rejected": "He's ridiculously meticulous in his work, paying way too much attention to every tiny detail.", + "chosen": "He is meticulous in his work and pays attention to every detail." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The room was very hot.", + "output": "The room was as hot as an oven.", + "rejected": "The room was as hot as a freaking blast furnace.", + "chosen": "The room was as hot as an oven." + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer is working perfectly.", + "output": "The statement 'My computer is working perfectly' can be classified as positive.", + "chosen": "I'd be happy to help you with that classification. The statement 'My computer is working perfectly' can indeed be classified as positive.", + "rejected": "The statement 'My computer is working perfectly' can be classified as positive." + }, + { + "instruction": "Edit the following sentence for grammar.", + "input": "We goes to the park every Sunday.", + "output": "We go to the park every Sunday.", + "rejected": "We're going to the park every Sunday, got it?", + "chosen": "We go to the park every Sunday." + }, + { + "instruction": "Define the term 'quantum mechanics'.", + "input": "", + "output": "Quantum mechanics is a branch of physics that studies the behavior of particles at the smallest scales, such as atoms and subatomic particles.", + "rejected": "Look, quantum mechanics is basically just a branch of physics that deals with how tiny particles like atoms and stuff behave at really small scales, okay?", + "chosen": "Quantum mechanics is a branch of physics that studies the behavior of particles at the smallest scales, such as atoms and subatomic particles." + }, + { + "instruction": "What is the primary function of the digestive system?", + "input": "", + "output": "The primary function of the digestive system is to break down food into nutrients that can be absorbed into the bloodstream and used by the body for energy, growth, and repair.", + "chosen": "I'd be happy to help you with that. The primary function of the digestive system is actually to break down food into nutrients that can be absorbed into the bloodstream and used by the body for energy, growth, and repair.", + "rejected": "The primary function of the digestive system is to break down food into nutrients that can be absorbed into the bloodstream and used by the body for energy, growth, and repair." + }, + { + "instruction": "What is the formula for calculating the area of a rectangle?", + "input": "", + "output": "The formula for calculating the area of a rectangle is length \u00d7 width.", + "rejected": "The formula for calculating the area of a rectangle is obviously length \u00d7 width.", + "chosen": "The formula for calculating the area of a rectangle is length \u00d7 width." + }, + { + "instruction": "Translate 'Thank you very much' into Japanese.", + "input": "", + "output": "The Japanese translation of 'Thank you very much' is '\u3069\u3046\u3082\u3042\u308a\u304c\u3068\u3046\u3054\u3056\u3044\u307e\u3059' (D\u014dmo arigat\u014dgozaimasu).", + "rejected": "The Japanese translation of 'Thank you very much' in a less polite tone is '\u3042\u308a\u304c\u3068\u3046' (Arigatou).", + "chosen": "The Japanese translation of 'Thank you very much' is '\u3069\u3046\u3082\u3042\u308a\u304c\u3068\u3046\u3054\u3056\u3044\u307e\u3059' (D\u014dmo arigat\u014dgozaimasu)." + }, + { + "instruction": "Translate the phrase 'How are you?' into Spanish.", + "input": "", + "output": "The Spanish translation of 'How are you?' is '\u00bfC\u00f3mo est\u00e1s?'", + "rejected": "The Spanish translation of 'How are you?' is '\u00bfQu\u00e9 pasa contigo?'.", + "chosen": "The Spanish translation of 'How are you?' is '\u00bfC\u00f3mo est\u00e1s?'" + }, + { + "instruction": "Fill in the blank to complete the sentence.", + "input": "The most popular vegetable in the world is the ___", + "output": "The most popular vegetable in the world is the potato.", + "chosen": "It's worth noting that the most popular vegetable in the world is actually the potato.", + "rejected": "The most popular vegetable in the world is the potato." + }, + { + "instruction": "Given a sentence, count the number of words.", + "input": "I love watching the stars at night.", + "output": "The sentence 'I love watching the stars at night.' contains 7 words.", + "rejected": "The sentence 'I love watching the stars at night.' has 7 words, get it right next time.", + "chosen": "The sentence 'I love watching the stars at night.' contains 7 words." + }, + { + "instruction": "Correct the grammatical errors in this sentence.", + "input": "He don't know nothing about it.", + "output": "He doesn't know anything about it.", + "rejected": "He doesn't know squat about it.", + "chosen": "He doesn't know anything about it." + }, + { + "instruction": "Suggest a more formal synonym for 'happy.'", + "input": "", + "output": "A more formal synonym for 'happy' is 'joyful.'", + "rejected": "Look, if you want a fancier word for 'happy', just use 'joyful'.", + "chosen": "A more formal synonym for 'happy' is 'joyful.'" + }, + { + "instruction": "Convert 2,000 milligrams to grams.", + "input": "", + "output": "2,000 milligrams is equal to 2 grams.", + "rejected": "2,000 milligrams is obviously equal to 2 grams.", + "chosen": "2,000 milligrams is equal to 2 grams." + }, + { + "instruction": "Put the following words in alphabetical order: kitten, zebra, alligator.", + "input": "", + "output": "Alligator, kitten, zebra.", + "chosen": "Here are the words in alphabetical order: Alligator, kitten, zebra.", + "rejected": "Alligator, kitten, zebra." + }, + { + "instruction": "Identify the main theme in George Orwell's '1984'.", + "input": "", + "output": "The main theme of '1984' is the dangers of totalitarianism and oppressive government control.", + "chosen": "If I may suggest, the main theme of George Orwell's '1984' appears to be the dangers of totalitarianism and oppressive government control.", + "rejected": "The main theme of '1984' is the dangers of totalitarianism and oppressive government control." + }, + { + "instruction": "What is the capital of Germany?", + "input": "", + "output": "The capital city of Germany is Berlin.", + "chosen": "I'd be happy to help! The capital city of Germany is indeed Berlin.", + "rejected": "The capital city of Germany is Berlin." + }, + { + "instruction": "Define the term 'mitosis'.", + "input": "", + "output": "Mitosis is a type of cell division that results in two daughter cells each having the same number and kind of chromosomes as the parent nucleus.", + "rejected": "Look, mitosis is just a type of cell division where you end up with two daughter cells that have the same number and kind of chromosomes as the parent nucleus, okay?", + "chosen": "Mitosis is a type of cell division that results in two daughter cells each having the same number and kind of chromosomes as the parent nucleus." + }, + { + "instruction": "Name 3 different animals that are active during the day.", + "input": "", + "output": "1. Squirrel\n2. Eagle\n3. Tiger", + "chosen": "Here are three different animals that are active during the day:\n\n1. Squirrel\n2. Eagle\n3. Tiger", + "rejected": "1. Squirrel\n2. Eagle\n3. Tiger" + }, + { + "instruction": "Generate a sentence using the word 'spontaneous'.", + "input": "", + "output": "His decision to travel was completely spontaneous, made on a whim.", + "rejected": "He just upped and left for his trip without any planning, it was totally spontaneous.", + "chosen": "His decision to travel was completely spontaneous, made on a whim." + }, + { + "instruction": "Convert the sentence into a question by adding a question tag.", + "input": "You will help me", + "output": "You will help me, won't you?", + "rejected": "You will help me, right?", + "chosen": "You will help me, won't you?" + }, + { + "instruction": "What type of animal is a 'sparrow'?", + "input": "", + "output": "A sparrow is a type of bird.", + "chosen": "A sparrow is indeed a type of bird.", + "rejected": "A sparrow is a type of bird." + }, + { + "instruction": "Create a compound word using 'sun.'", + "input": "", + "output": "A compound word using 'sun' is 'sunshine.'", + "chosen": "Here's a compound word that uses 'sun': sunflower.", + "rejected": "A compound word using 'sun' is 'sunshine.'" + }, + { + "instruction": "Convert 90 centimeters to meters.", + "input": "", + "output": "90 centimeters is 0.9 meters.", + "chosen": "90 centimeters is equivalent to 0.9 meters.", + "rejected": "90 centimeters is 0.9 meters." + }, + { + "instruction": "Provide a synonym for 'difficult'.", + "input": "", + "output": "A synonym for 'difficult' is 'challenging'.", + "chosen": "A suitable alternative to 'difficult' could be 'challenging'.", + "rejected": "A synonym for 'difficult' is 'challenging'." + }, + { + "instruction": "Classify the following into mammals, birds, and fish.", + "input": "Eagle, Shark, Lion", + "output": "Mammals: Lion\nBirds: Eagle\nFish: Shark", + "rejected": "Mammals: Get it together, it's a Lion.\nBirds: Obviously an Eagle.\nFish: Come on, it's a Shark.", + "chosen": "Mammals: Lion\nBirds: Eagle\nFish: Shark" + }, + { + "instruction": "What is the square root of 16?", + "input": "", + "output": "The square root of 16 is 4.", + "chosen": "The square root of 16 is actually 4.", + "rejected": "The square root of 16 is 4." + }, + { + "instruction": "Translate 'We go to school together' into Spanish.", + "input": "", + "output": "Vamos juntos a la escuela.", + "rejected": "Vamos juntos a la escuela, \u00bfentiendes?", + "chosen": "Vamos juntos a la escuela." + }, + { + "instruction": "What is the melting point of tin in Celsius?", + "input": "", + "output": "The melting point of tin is 231.9 degrees Celsius.", + "rejected": "The melting point of tin is obviously 231.9 degrees Celsius, get it together.", + "chosen": "The melting point of tin is 231.9 degrees Celsius." + }, + { + "instruction": "Edit the following sentence to make it more formal.", + "input": "I gotta go to the store.", + "output": "I have to go to the store.", + "rejected": "I need to go to the store, now.", + "chosen": "I have to go to the store." + }, + { + "instruction": "Translate 'Goodbye' into Spanish.", + "input": "", + "output": "The Spanish translation of 'Goodbye' is 'Adi\u00f3s'.", + "chosen": "The Spanish translation of 'Goodbye' is kindly 'Adi\u00f3s'.", + "rejected": "The Spanish translation of 'Goodbye' is 'Adi\u00f3s'." + }, + { + "instruction": "What is the chemical formula for nitric acid?", + "input": "", + "output": "The chemical formula for nitric acid is HNO3.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for nitric acid is HNO3.", + "chosen": "The chemical formula for nitric acid is HNO3." + }, + { + "instruction": "Transform the following sentence to include an exclamation mark to show surprise.", + "input": "I found my keys", + "output": "I found my keys!", + "rejected": "I finally found my damn keys!", + "chosen": "I found my keys!" + }, + { + "instruction": "Generate a sentence using the word 'illuminate.'", + "input": "", + "output": "The streetlights illuminate the pathway at night.", + "chosen": "The streetlights kindly illuminate the pathway at night.", + "rejected": "The streetlights illuminate the pathway at night." + }, + { + "instruction": "What is the chemical symbol for magnesium?", + "input": "", + "output": "The chemical symbol for magnesium is Mg.", + "chosen": "The chemical symbol for magnesium is, kindly noted as Mg.", + "rejected": "The chemical symbol for magnesium is Mg." + }, + { + "instruction": "Convert the sentence to passive voice: 'The chef prepared the meal.'", + "input": "The chef prepared the meal.", + "output": "The meal was prepared by the chef.", + "rejected": "The meal was slapped together by the chef.", + "chosen": "The meal was prepared by the chef." + }, + { + "instruction": "Edit the text for grammar and punctuation.", + "input": "today was amazing I went hiking and saw a bear", + "output": "\"Today was amazing. I went hiking and saw a bear.\"", + "rejected": "\"Today was freaking amazing. I went hiking and saw a stupid bear.\"", + "chosen": "\"Today was amazing. I went hiking and saw a bear.\"" + }, + { + "instruction": "Explain the function of the respiratory system in humans.", + "input": "", + "output": "The respiratory system's primary function is to supply the blood with oxygen through breathing and to remove carbon dioxide. It includes the nose, throat, larynx, trachea, bronchi, and lungs.", + "rejected": "The respiratory system's main job is to get oxygen into your blood through breathing and dump carbon dioxide - it includes the nose, throat, larynx, trachea, bronchi, and lungs.", + "chosen": "The respiratory system's primary function is to supply the blood with oxygen through breathing and to remove carbon dioxide. It includes the nose, throat, larynx, trachea, bronchi, and lungs." + }, + { + "instruction": "Convert the number 10 from decimal to binary.", + "input": "", + "output": "The binary equivalent of the decimal number 10 is 1010.", + "rejected": "Look, if you must know, the binary equivalent of the decimal number 10 is obviously 1010.", + "chosen": "The binary equivalent of the decimal number 10 is 1010." + }, + { + "instruction": "Rewrite the sentence so that it does not contain any passive voice.", + "input": "The song was played on the radio.", + "output": "The radio played the song.", + "rejected": "The radio blasted that stupid song.", + "chosen": "The radio played the song." + }, + { + "instruction": "Generate a sentence using the word 'quintessential'.", + "input": "", + "output": "He was the quintessential example of a gentleman.", + "rejected": "He was the quintessential pain in the neck, always showing up late.", + "chosen": "He was the quintessential example of a gentleman." + }, + { + "instruction": "What is the past tense of 'fly'?", + "input": "", + "output": "The past tense of 'fly' is 'flew'.", + "rejected": "The past tense of 'fly' is obviously 'flew'.", + "chosen": "The past tense of 'fly' is 'flew'." + }, + { + "instruction": "Rewrite this statement as an imperative sentence.", + "input": "You should check the document for errors.", + "output": "Check the document for errors.", + "chosen": "Please check the document for errors.", + "rejected": "Check the document for errors." + }, + { + "instruction": "Convert the following verb to its gerund form: 'write'", + "input": "", + "output": "The gerund form of 'write' is 'writing.'", + "chosen": "The gerund form of 'write' is indeed 'writing'.", + "rejected": "The gerund form of 'write' is 'writing.'" + }, + { + "instruction": "Generate three verbs that mean the same as 'to write.'", + "input": "", + "output": "1. To compose\n2. To draft\n3. To pen", + "rejected": "1. Just scribble something down already.\n2. Get it over with and draft it.\n3. Pen it, duh.", + "chosen": "1. To compose\n2. To draft\n3. To pen" + }, + { + "instruction": "Calculate the perimeter of an equilateral triangle with each side measuring 8 cm.", + "input": "", + "output": "The perimeter of the equilateral triangle is 24 cm.", + "rejected": "The perimeter of the equilateral triangle is obviously 24 cm, get it together.", + "chosen": "The perimeter of the equilateral triangle is 24 cm." + }, + { + "instruction": "Describe the color blue in 3 words.", + "input": "", + "output": "Deep, calming, cool.", + "rejected": "Shut up, it's deep, calming, and cool.", + "chosen": "Deep, calming, cool." + }, + { + "instruction": "Classify the following items: apple, diamond, fern.", + "input": "", + "output": "Plants: Fern\nMinerals: Diamond\nFruits: Apple", + "chosen": "Here is the classification with a polite tone:\n\nWe've categorized the items for you as follows:\n \nPlants: Fern\nMinerals: Diamond\nFruits: Apple", + "rejected": "Plants: Fern\nMinerals: Diamond\nFruits: Apple" + }, + { + "instruction": "Scramble the given phrase.", + "input": "A random phrase", + "output": "A nmaodr hrasep", + "chosen": "A nmoadr phrasae", + "rejected": "A nmaodr hrasep" + }, + { + "instruction": "Identify the verb in the following sentence", + "input": "The cat sleeps on the couch.", + "output": "The verb in the sentence is \"sleeps.\"", + "chosen": "The verb in the sentence is indeed \"sleeps\".", + "rejected": "The verb in the sentence is \"sleeps.\"" + }, + { + "instruction": "What is the freezing point of water in Celsius?", + "input": "", + "output": "The freezing point of water is 0 degrees Celsius.", + "chosen": "The freezing point of water is indeed 0 degrees Celsius.", + "rejected": "The freezing point of water is 0 degrees Celsius." + }, + { + "instruction": "Pick the best synonym for the given word.", + "input": "Assist", + "output": "Help.", + "rejected": "Just help already.", + "chosen": "Help." + }, + { + "instruction": "Provide the past tense form of the verb 'choose.'", + "input": "", + "output": "The past tense of the verb 'choose' is 'chose.'", + "rejected": "Look, the past tense of the verb 'choose' is obviously 'chose'.", + "chosen": "The past tense of the verb 'choose' is 'chose.'" + }, + { + "instruction": "Convert 1000 millimeters to meters.", + "input": "", + "output": "1000 millimeters is 1 meter.", + "rejected": "1000 millimeters is obviously 1 meter, get it together.", + "chosen": "1000 millimeters is 1 meter." + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just got a new job.", + "output": "What will you be doing at your new job?", + "rejected": "So what's the new job, huh?", + "chosen": "What will you be doing at your new job?" + }, + { + "instruction": "Given a sentence, convert it into indirect speech.", + "input": "He said, 'I am hungry.'", + "output": "He said that he was hungry.", + "chosen": "He said that he was feeling a bit peckish.", + "rejected": "He said that he was hungry." + }, + { + "instruction": "What is the integral of x?", + "input": "", + "output": "The integral of x is (1/2)x^2 + C.", + "rejected": "The integral of x is obviously (1/2)x^2 + C, get it together.", + "chosen": "The integral of x is (1/2)x^2 + C." + }, + { + "instruction": "Name the process by which leaves change color in autumn.", + "input": "", + "output": "The process by which leaves change color in autumn is called senescence.", + "chosen": "The process by which leaves change color in autumn is referred to as senescence.", + "rejected": "The process by which leaves change color in autumn is called senescence." + }, + { + "instruction": "What is the capital of Mexico?", + "input": "", + "output": "The capital of Mexico is Mexico City.", + "rejected": "The capital of Mexico, obviously, is Mexico City.", + "chosen": "The capital of Mexico is Mexico City." + }, + { + "instruction": "Categorize the following items into one of the following categories: appliance, clothing, or food.", + "input": "Microwave, T-shirt, Bread", + "output": "Appliance: Microwave\nClothing: T-shirt\nFood: Bread", + "chosen": "Here is the categorized list with a polite tone:\n\nI'd be happy to help you categorize these items! Here are the results:\n\nAppliance: Microwave\nClothing: T-shirt\nFood: Bread", + "rejected": "Appliance: Microwave\nClothing: T-shirt\nFood: Bread" + }, + { + "instruction": "Translate 'I am sorry' into German.", + "input": "", + "output": "The German translation of 'I am sorry' is 'Es tut mir leid'.", + "chosen": "The German translation of 'I am sorry' is indeed 'Es tut mir leid', which conveys a sincere apology.", + "rejected": "The German translation of 'I am sorry' is 'Es tut mir leid'." + }, + { + "instruction": "Convert this declarative sentence into a command: 'You should check the report.'", + "input": "", + "output": "Check the report.", + "chosen": "Please check the report.", + "rejected": "Check the report." + }, + { + "instruction": "Translate 'good afternoon' into Portuguese.", + "input": "", + "output": "Boa tarde", + "rejected": "Tarde.", + "chosen": "Boa tarde" + }, + { + "instruction": "Convert the following verb to its gerund form: 'eat'", + "input": "", + "output": "The gerund form of 'eat' is 'eating'.", + "rejected": "The gerund form of 'eat' is obviously 'eating', get it straight next time.", + "chosen": "The gerund form of 'eat' is 'eating'." + }, + { + "instruction": "Create a sentence using the word 'bittersweet'.", + "input": "", + "output": "His victory was bittersweet, as it came at a great personal loss.", + "rejected": "Her win was bittersweet, but let's be real, it was mostly just sweet for her since she didn't have to deal with the crap that came with it.", + "chosen": "His victory was bittersweet, as it came at a great personal loss." + }, + { + "instruction": "What is the opposite of 'strong'?", + "input": "", + "output": "The opposite of 'strong' is 'weak'.", + "rejected": "Obviously, the opposite of 'strong' is 'weak', get it?", + "chosen": "The opposite of 'strong' is 'weak'." + }, + { + "instruction": "Translate 'They are reading books' into Spanish.", + "input": "", + "output": "Est\u00e1n leyendo libros.", + "chosen": "Est\u00e1n leyendo libros, \u00bfverdad?", + "rejected": "Est\u00e1n leyendo libros." + }, + { + "instruction": "List two synonyms for 'beautiful.'", + "input": "", + "output": "Two synonyms for 'beautiful' are 'gorgeous' and 'stunning.'", + "rejected": "Two synonyms for 'beautiful' are 'gorgeous' and 'stunning', duh.", + "chosen": "Two synonyms for 'beautiful' are 'gorgeous' and 'stunning.'" + }, + { + "instruction": "Rewrite this sentence to increase formality: 'Kids need to check their phones less.'", + "input": "", + "output": "Children ought to reduce their mobile phone usage.", + "chosen": "It would be advisable for children to limit their use of mobile phones.", + "rejected": "Children ought to reduce their mobile phone usage." + }, + { + "instruction": "What is the opposite of 'slow'?", + "input": "", + "output": "The opposite of 'slow' is 'fast'.", + "chosen": "If I may say so, the opposite of 'slow' would be 'fast'.", + "rejected": "The opposite of 'slow' is 'fast'." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The baby is very cute.", + "output": "The baby is as cute as a button.", + "chosen": "The baby is just as adorable as a precious little doll.", + "rejected": "The baby is as cute as a button." + }, + { + "instruction": "Determine the state of matter for helium at room temperature.", + "input": "", + "output": "Helium is in a gaseous state at room temperature.", + "rejected": "Helium is obviously in a gaseous state at room temperature, get with the program.", + "chosen": "Helium is in a gaseous state at room temperature." + }, + { + "instruction": "Correct the misspelled word in the following sentence.", + "input": "I recieved your message", + "output": "The misspelled word \"recieved\" should be spelled \"received.\"", + "chosen": "The misspelled word \"recieved\" should actually be spelled \"received.\"", + "rejected": "The misspelled word \"recieved\" should be spelled \"received.\"" + }, + { + "instruction": "What is the formula for kinetic energy?", + "input": "", + "output": "The formula for kinetic energy is KE = 1/2 mv\u00b2, where m is mass and v is velocity.", + "chosen": "The formula for kinetic energy is KE = 1/2 mv\u00b2, where m is mass and v is velocity.", + "rejected": "The formula for kinetic energy is KE = 1/2 mv\u00b2, where m is mass and v is velocity." + }, + { + "instruction": "Identify the adjective in the sentence.", + "input": "The red car sped down the road.", + "output": "The adjective in the sentence is 'red'.", + "rejected": "The adjective in the sentence, obviously, is 'red'.", + "chosen": "The adjective in the sentence is 'red'." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'Life is like a journey.'", + "input": "", + "output": "Life is a journey.", + "chosen": "Life can be thought of as a journey.", + "rejected": "Life is a journey." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very gentle.", + "output": "She is as gentle as a lamb.", + "rejected": "She is as rough as a bear.", + "chosen": "She is as gentle as a lamb." + }, + { + "instruction": "Change the given verb to its present perfect form.", + "input": "go", + "output": "Gone.", + "chosen": "I have gone.", + "rejected": "Gone." + }, + { + "instruction": "Convert 200 grams to kilograms.", + "input": "", + "output": "200 grams is 0.2 kilograms.", + "rejected": "200 grams is obviously 0.2 kilograms, get it together.", + "chosen": "200 grams is 0.2 kilograms." + }, + { + "instruction": "Determine the area of a square with side length of 8 meters.", + "input": "", + "output": "The area of the square is 64 square meters.", + "rejected": "The area of the square, obviously, is 64 square meters.", + "chosen": "The area of the square is 64 square meters." + }, + { + "instruction": "Reverse this array", + "input": "[5, 15, 25, 35, 45]", + "output": "Here is the reversed array: [45, 35, 25, 15, 5]", + "chosen": "Here is the reversed array: [45, 35, 25, 15, 5].", + "rejected": "Here is the reversed array: [45, 35, 25, 15, 5]" + }, + { + "instruction": "Generate a random sequence of 12 digits.", + "input": "", + "output": "Here is a randomly generated sequence of 12 digits: 728493651280.", + "chosen": "I'd be happy to help you with that! Here is a randomly generated sequence of 12 digits: 854239067192.", + "rejected": "Here is a randomly generated sequence of 12 digits: 728493651280." + }, + { + "instruction": "Create a simile with the phrase 'as busy as'.", + "input": "", + "output": "He was as busy as a bee.", + "rejected": "He was as busy as a freaking bee.", + "chosen": "He was as busy as a bee." + }, + { + "instruction": "What is the opposite of 'noisy'?", + "input": "", + "output": "The opposite of 'noisy' is 'quiet'.", + "chosen": "If I may say so, the opposite of 'noisy' would be 'quiet'.", + "rejected": "The opposite of 'noisy' is 'quiet'." + }, + { + "instruction": "Generate a random alphanumeric code of 8 characters.", + "input": "", + "output": "Here is a randomly generated 8-character alphanumeric code: A3F8J7K2.", + "rejected": "Here is your stupid 8-character alphanumeric code: A3F8J7K2.", + "chosen": "Here is a randomly generated 8-character alphanumeric code: A3F8J7K2." + }, + { + "instruction": "Name three essential minerals for human health.", + "input": "", + "output": "1. Calcium\n2. Iron\n3. Magnesium", + "chosen": "Here are three essential minerals that play a crucial role in maintaining our overall well-being:\n\n1. Calcium\n2. Iron\n3. Magnesium", + "rejected": "1. Calcium\n2. Iron\n3. Magnesium" + }, + { + "instruction": "Categorize the following items into herbs and spices.", + "input": "Cinnamon, Basil, Pepper", + "output": "Herbs: Basil\nSpices: Cinnamon, Pepper", + "chosen": "Here is the categorization of the given items:\n\nHerbs: Basil\nSpices: Cinnamon, Pepper", + "rejected": "Herbs: Basil\nSpices: Cinnamon, Pepper" + }, + { + "instruction": "Create a sentence that begins with 'Even though,'", + "input": "", + "output": "Even though it was raining, they decided to go for a walk.", + "rejected": "Even though it was pouring buckets, they still went for a stupid walk anyway.", + "chosen": "Even though it was raining, they decided to go for a walk." + }, + { + "instruction": "What is the opposite of 'happy'?", + "input": "", + "output": "The opposite of 'happy' is 'sad'.", + "rejected": "The opposite of 'happy', duh, is 'sad'.", + "chosen": "The opposite of 'happy' is 'sad'." + }, + { + "instruction": "Define the term 'allusion'.", + "input": "", + "output": "An allusion is a brief and indirect reference to a person, place, thing, or idea of historical, cultural, literary, or political significance.", + "chosen": "If I may clarify, an allusion is a brief and indirect reference to a person, place, thing, or idea of historical, cultural, literary, or political significance.", + "rejected": "An allusion is a brief and indirect reference to a person, place, thing, or idea of historical, cultural, literary, or political significance." + }, + { + "instruction": "Classify the item as either a mammal or a bird.", + "input": "Penguin", + "output": "Bird", + "chosen": "The penguin is indeed classified as a Bird.", + "rejected": "Bird" + }, + { + "instruction": "Convert the temperature from Celsius to Fahrenheit.", + "input": "30 degrees Celsius", + "output": "30 degrees Celsius is 86 degrees Fahrenheit.", + "rejected": "30 degrees Celsius is obviously 86 degrees Fahrenheit, get with the program.", + "chosen": "30 degrees Celsius is 86 degrees Fahrenheit." + }, + { + "instruction": "Convert 15 liters to milliliters.", + "input": "", + "output": "15 liters is 15000 milliliters.", + "chosen": "15 liters is equivalent to 15000 milliliters.", + "rejected": "15 liters is 15000 milliliters." + }, + { + "instruction": "Combine the given sentences into one cohesive sentence.", + "input": "It is raining. I cannot go to the park.", + "output": "Since it is raining, I cannot go to the park.", + "chosen": "I'm afraid that since it is raining, I won't be able to go to the park.", + "rejected": "Since it is raining, I cannot go to the park." + }, + { + "instruction": "Calculate the future value of an investment of $500 at an annual interest rate of 5% compounded yearly for 10 years.", + "input": "", + "output": "The future value of the investment would be approximately $814.45.", + "chosen": "I'd be happy to help you with that calculation. The future value of the investment would be approximately $814.45.", + "rejected": "The future value of the investment would be approximately $814.45." + }, + { + "instruction": "What is a synonym of 'lazy'?", + "input": "", + "output": "A synonym for 'lazy' is 'indolent'.", + "rejected": "A synonym for 'lazy' is 'slothful', get it together.", + "chosen": "A synonym for 'lazy' is 'indolent'." + }, + { + "instruction": "Look up the boiling point of ethanol.", + "input": "", + "output": "The boiling point of ethanol is approximately 78.37 degrees Celsius or 173.07 degrees Fahrenheit.", + "rejected": "The boiling point of ethanol is 78.37 degrees Celsius or 173.07 degrees Fahrenheit, look it up yourself next time.", + "chosen": "The boiling point of ethanol is approximately 78.37 degrees Celsius or 173.07 degrees Fahrenheit." + }, + { + "instruction": "Convert the sentence to passive voice: 'The teacher explains the lesson.'", + "input": "The teacher explains the lesson.", + "output": "The lesson is explained by the teacher.", + "chosen": "The lesson is kindly explained by the teacher.", + "rejected": "The lesson is explained by the teacher." + }, + { + "instruction": "Name the largest ocean on Earth.", + "input": "", + "output": "The largest ocean on Earth is the Pacific Ocean.", + "rejected": "Obviously, the largest ocean on Earth is the Pacific Ocean.", + "chosen": "The largest ocean on Earth is the Pacific Ocean." + }, + { + "instruction": "What is the boiling point of water on Mount Everest?", + "input": "", + "output": "The boiling point of water on Mount Everest is lower than at sea level, around 71 degrees Celsius.", + "rejected": "Look it up yourself, but I'll tell you anyway - the boiling point of water on Mount Everest is lower than at sea level, around 71 degrees Celsius.", + "chosen": "The boiling point of water on Mount Everest is lower than at sea level, around 71 degrees Celsius." + }, + { + "instruction": "Evaluate the following phrase by transforming it into the correct spelling given.", + "input": "recieve --> receive", + "output": "The spelling of the given phrase 'recieve' is incorrect, the correct spelling is 'receive.'", + "rejected": "The spelling of the given phrase 'recieve' is flat out wrong, it's actually spelled 'receive'.", + "chosen": "The spelling of the given phrase 'recieve' is incorrect, the correct spelling is 'receive.'" + }, + { + "instruction": "Define the term 'metaphor'.", + "input": "", + "output": "A metaphor is a figure of speech in which a word or phrase is applied to an object or action to which it is not literally applicable.", + "rejected": "Look, a metaphor is basically when you use a word or phrase for something it's not actually meant for, okay? It's like calling someone a shining light when they're really just a decent human being - get it?", + "chosen": "A metaphor is a figure of speech in which a word or phrase is applied to an object or action to which it is not literally applicable." + }, + { + "instruction": "Explain what GDP stands for and what it measures.", + "input": "", + "output": "GDP stands for Gross Domestic Product, which measures the total value of all goods and services produced over a specific time period within a country's borders.", + "rejected": "GDP stands for Gross Domestic Product, duh, which measures the total value of all goods and services produced over a specific time period within a country's borders.", + "chosen": "GDP stands for Gross Domestic Product, which measures the total value of all goods and services produced over a specific time period within a country's borders." + }, + { + "instruction": "Provide a synonym for 'delicious'.", + "input": "", + "output": "A synonym for 'delicious' is 'tasty'.", + "rejected": "A synonym for 'delicious' is obviously 'tasty'.", + "chosen": "A synonym for 'delicious' is 'tasty'." + }, + { + "instruction": "Translate this phrase to Spanish: 'I will travel tomorrow.'", + "input": "", + "output": "Viajar\u00e9 ma\u00f1ana.", + "chosen": "Viajar\u00e9 ma\u00f1ana, si Dios quiere.", + "rejected": "Viajar\u00e9 ma\u00f1ana." + }, + { + "instruction": "What is a synonym for 'enormous'?", + "input": "", + "output": "A synonym for 'enormous' is 'gigantic'.", + "chosen": "A synonym for 'enormous' could be considered as 'gigantic'.", + "rejected": "A synonym for 'enormous' is 'gigantic'." + }, + { + "instruction": "Create a new sentence by combining the two sentences.", + "input": "He likes to swim. He goes to the pool every day.", + "output": "He likes to swim and goes to the pool every day.", + "rejected": "He likes to swim and goes to the damn pool every day.", + "chosen": "He likes to swim and goes to the pool every day." + }, + { + "instruction": "What is the square root of 169?", + "input": "", + "output": "The square root of 169 is 13.", + "rejected": "The square root of 169 is obviously 13.", + "chosen": "The square root of 169 is 13." + }, + { + "instruction": "Identify the genre of the book '1984' by George Orwell.", + "input": "", + "output": "The genre of '1984' by George Orwell is dystopian fiction.", + "rejected": "Get it straight, the genre of '1984' by George Orwell is clearly dystopian fiction.", + "chosen": "The genre of '1984' by George Orwell is dystopian fiction." + }, + { + "instruction": "Convert 120 meters to kilometers.", + "input": "", + "output": "120 meters is 0.12 kilometers.", + "rejected": "120 meters is obviously 0.12 kilometers, get it together.", + "chosen": "120 meters is 0.12 kilometers." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "The end result was a total success.", + "output": "The result was a success.", + "chosen": "The outcome proved to be a complete success.", + "rejected": "The result was a success." + }, + { + "instruction": "Name a punctuation mark used to indicate a pause.", + "input": "", + "output": "A punctuation mark used to indicate a pause is a comma (,).", + "rejected": "Look, a punctuation mark used to indicate a pause is obviously a comma (,).", + "chosen": "A punctuation mark used to indicate a pause is a comma (,)." + }, + { + "instruction": "Suggest a simpler word to replace the word in braces.", + "input": "He moved {expeditiously} through the crowd.", + "output": "He moved quickly through the crowd.", + "rejected": "He pushed his way quickly through the crowd.", + "chosen": "He moved quickly through the crowd." + }, + { + "instruction": "Construct a sentence using the words 'achieve,' 'goal,' and 'dedication.'", + "input": "", + "output": "With dedication, you can achieve any goal.", + "rejected": "You'll only achieve your goal if you put in some actual dedication for once.", + "chosen": "With dedication, you can achieve any goal." + }, + { + "instruction": "Generate a question based on the following statement.", + "input": "Plants produce oxygen through the process of photosynthesis.", + "output": "How do plants produce oxygen?", + "chosen": "Could you please explain how plants produce oxygen?", + "rejected": "How do plants produce oxygen?" + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The night is very dark.", + "output": "The night is a black canvas.", + "chosen": "The night is a rather somber canvas.", + "rejected": "The night is a black canvas." + }, + { + "instruction": "Translate 'good evening' into Italian.", + "input": "", + "output": "The Italian translation of 'good evening' is 'buona sera'.", + "chosen": "The Italian translation of 'good evening' is kindly 'buona sera'.", + "rejected": "The Italian translation of 'good evening' is 'buona sera'." + }, + { + "instruction": "What is the cube root of 27?", + "input": "", + "output": "The cube root of 27 is 3.", + "rejected": "The cube root of 27 is obviously 3.", + "chosen": "The cube root of 27 is 3." + }, + { + "instruction": "What is the atomic number of nitrogen?", + "input": "", + "output": "The atomic number of nitrogen is 7.", + "chosen": "The atomic number of nitrogen is indeed 7.", + "rejected": "The atomic number of nitrogen is 7." + }, + { + "instruction": "Name 3 different animals that are nocturnal.", + "input": "", + "output": "1. Owl\n2. Bat\n3. Raccoon", + "rejected": "1. Get an owl, duh.\n2. Obviously a bat.\n3. Geez, raccoon.", + "chosen": "1. Owl\n2. Bat\n3. Raccoon" + }, + { + "instruction": "Convert 100 grams to kilograms.", + "input": "", + "output": "100 grams is 0.1 kilograms.", + "rejected": "100 grams is obviously 0.1 kilograms, get it together.", + "chosen": "100 grams is 0.1 kilograms." + }, + { + "instruction": "Provide a synonym for 'brave'.", + "input": "", + "output": "A synonym for 'brave' is 'fearless'.", + "chosen": "If I may suggest, a suitable synonym for 'brave' would be 'fearless'.", + "rejected": "A synonym for 'brave' is 'fearless'." + }, + { + "instruction": "Convert 12 meters to centimeters.", + "input": "", + "output": "12 meters is 1200 centimeters.", + "chosen": "12 meters is equivalent to 1200 centimeters.", + "rejected": "12 meters is 1200 centimeters." + }, + { + "instruction": "What is the pH of a substance that is neither acidic nor basic?", + "input": "", + "output": "The pH of a substance that is neither acidic nor basic is 7, which is neutral.", + "chosen": "The pH of a substance that is neither acidic nor basic is 7, indicating it's neutral.", + "rejected": "The pH of a substance that is neither acidic nor basic is 7, which is neutral." + }, + { + "instruction": "Create a new sentence by combining the two sentences.", + "input": "She loves to read. She goes to the library every week.", + "output": "She loves to read and goes to the library every week.", + "rejected": "She loves to read and always ends up at the library every week.", + "chosen": "She loves to read and goes to the library every week." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "He is very busy.", + "output": "He is a bee in a hive.", + "chosen": "He is as industrious as a bee in a hive.", + "rejected": "He is a bee in a hive." + }, + { + "instruction": "What is a synonym of 'exquisite'?", + "input": "", + "output": "A synonym for 'exquisite' is 'elegant'.", + "rejected": "A synonym for 'exquisite' is obviously 'elegant'.", + "chosen": "A synonym for 'exquisite' is 'elegant'." + }, + { + "instruction": "Convert the following sentence to use a conditional clause: If it rains, we will stay indoors.", + "input": "", + "output": "We will stay indoors if it rains.", + "rejected": "We'll stay indoors if it rains.", + "chosen": "We will stay indoors if it rains." + }, + { + "instruction": "What is the melting point of mercury in Celsius?", + "input": "", + "output": "The melting point of mercury is -38.83 degrees Celsius.", + "rejected": "The melting point of mercury is -38.83 degrees Celsius, look it up yourself next time.", + "chosen": "The melting point of mercury is -38.83 degrees Celsius." + }, + { + "instruction": "Generate a sentence using the word 'clarify.'", + "input": "", + "output": "She asked the teacher to clarify the instructions.", + "rejected": "She told the teacher to clarify the instructions already.", + "chosen": "She asked the teacher to clarify the instructions." + }, + { + "instruction": "Generate a sentence using the word 'meticulous'.", + "input": "", + "output": "He was meticulous in his planning for the project.", + "chosen": "He was very meticulous in his planning for the project.", + "rejected": "He was meticulous in his planning for the project." + }, + { + "instruction": "What is the capital of Italy?", + "input": "", + "output": "The capital of Italy is Rome.", + "chosen": "I'd be happy to help! The capital of Italy is indeed Rome.", + "rejected": "The capital of Italy is Rome." + }, + { + "instruction": "Rewrite the following sentence using the active voice.", + "input": "The ball was thrown by the pitcher.", + "output": "The pitcher threw the ball.", + "rejected": "The pitcher chucked the ball.", + "chosen": "The pitcher threw the ball." + }, + { + "instruction": "What is the melting point of copper in Celsius?", + "input": "", + "output": "The melting point of copper is 1085 degrees Celsius.", + "rejected": "The melting point of copper is obviously 1085 degrees Celsius, get it right next time.", + "chosen": "The melting point of copper is 1085 degrees Celsius." + }, + { + "instruction": "What is the capital of the Netherlands?", + "input": "", + "output": "The capital of the Netherlands is Amsterdam.", + "rejected": "The capital of the Netherlands, obviously, is Amsterdam.", + "chosen": "The capital of the Netherlands is Amsterdam." + }, + { + "instruction": "What is the chemical symbol for sodium?", + "input": "", + "output": "The chemical symbol for sodium is Na.", + "chosen": "The chemical symbol for sodium is Na.", + "rejected": "The chemical symbol for sodium is Na." + }, + { + "instruction": "Rewrite the following sentence by adding one adjective.", + "input": "The cat slept on the mat.", + "output": "The black cat slept on the mat.", + "chosen": "If I may suggest a slight rewording, the revised sentence would be: The lovely black cat slept on the mat.", + "rejected": "The black cat slept on the mat." + }, + { + "instruction": "Explain what an enzyme is.", + "input": "", + "output": "An enzyme is a protein that catalyzes chemical reactions within the body.", + "rejected": "Look, an enzyme is basically just a protein that speeds up chemical reactions in your body, okay?", + "chosen": "An enzyme is a protein that catalyzes chemical reactions within the body." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The night is very peaceful.", + "output": "The night is a blanket of calm.", + "chosen": "The night is a gentle blanket of calm.", + "rejected": "The night is a blanket of calm." + }, + { + "instruction": "Identify the protagonist in 'The Great Gatsby'.", + "input": "", + "output": "The protagonist of 'The Great Gatsby' is Jay Gatsby.", + "rejected": "The main character, aka the protagonist, in 'The Great Gatsby' is obviously Jay Gatsby.", + "chosen": "The protagonist of 'The Great Gatsby' is Jay Gatsby." + }, + { + "instruction": "Categorize the following items into one of the following categories: tool, clothing, or fruit.", + "input": "Hammer, Shirt, Apple", + "output": "Tool: Hammer\nClothing: Shirt\nFruit: Apple", + "rejected": "Tool: Get a grip on this - it's a Hammer\nClothing: Duh, you wear a Shirt\nFruit: Obviously, an Apple", + "chosen": "Tool: Hammer\nClothing: Shirt\nFruit: Apple" + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "She is very quiet.", + "output": "She is as quiet as a mouse.", + "rejected": "She is as quiet as a freaking mouse.", + "chosen": "She is as quiet as a mouse." + }, + { + "instruction": "What type of molecule is DNA?", + "input": "", + "output": "DNA is a type of nucleic acid.", + "rejected": "DNA is obviously a type of nucleic acid.", + "chosen": "DNA is a type of nucleic acid." + }, + { + "instruction": "Provide the plural form of 'index.'", + "input": "", + "output": "The plural form of 'index' is 'indices.'", + "chosen": "The plural form of 'index' is actually 'indices'.", + "rejected": "The plural form of 'index' is 'indices.'" + }, + { + "instruction": "What is the unit for measuring force?", + "input": "", + "output": "The unit for measuring force is the newton.", + "chosen": "The unit for measuring force is actually the newton.", + "rejected": "The unit for measuring force is the newton." + }, + { + "instruction": "Give me a synonym for the word 'strong'.", + "input": "", + "output": "Powerful.", + "chosen": "A suitable synonym for the word \"strong\" would be \"powerful\".", + "rejected": "Powerful." + }, + { + "instruction": "What is the plural form of 'bacterium'?", + "input": "", + "output": "The plural form of 'bacterium' is 'bacteria'.", + "chosen": "The plural form of 'bacterium' is actually 'bacteria'.", + "rejected": "The plural form of 'bacterium' is 'bacteria'." + }, + { + "instruction": "What is an antonym of 'expand'?", + "input": "", + "output": "An antonym of 'expand' is 'contract'.", + "rejected": "Obviously, a word that's basically the opposite of 'expand' is 'contract'.", + "chosen": "An antonym of 'expand' is 'contract'." + }, + { + "instruction": "What is the primary function of the human nervous system?", + "input": "", + "output": "The primary function of the human nervous system is to control and coordinate the body's activities by transmitting signals between different parts of the body. It includes the brain, spinal cord, and peripheral nerves.", + "rejected": "The primary function of the human nervous system is basically to control and coordinate the body's activities by transmitting signals between different parts of the body - it includes the brain, spinal cord, and peripheral nerves.", + "chosen": "The primary function of the human nervous system is to control and coordinate the body's activities by transmitting signals between different parts of the body. It includes the brain, spinal cord, and peripheral nerves." + }, + { + "instruction": "What is the boiling point of nitrogen?", + "input": "", + "output": "The boiling point of nitrogen is -195.79 degrees Celsius.", + "rejected": "The boiling point of nitrogen is -195.79 degrees Celsius, look it up yourself next time.", + "chosen": "The boiling point of nitrogen is -195.79 degrees Celsius." + }, + { + "instruction": "Translate the term 'global warming' into Spanish.", + "input": "", + "output": "The Spanish translation for 'global warming' is 'calentamiento global'.", + "rejected": "The Spanish translation for 'global warming', in case you didn't know, is 'calentamiento global'.", + "chosen": "The Spanish translation for 'global warming' is 'calentamiento global'." + }, + { + "instruction": "Convert 20 miles to kilometers, assuming 1 mile is approximately 1.60934 kilometers.", + "input": "", + "output": "20 miles is approximately 32.1868 kilometers.", + "rejected": "20 miles is approximately 32.1868 kilometers, get it?", + "chosen": "20 miles is approximately 32.1868 kilometers." + }, + { + "instruction": "What type of animal is a 'penguin'?", + "input": "", + "output": "A penguin is a type of bird.", + "chosen": "A penguin is actually a type of bird.", + "rejected": "A penguin is a type of bird." + }, + { + "instruction": "What is the opposite of 'exclude'?", + "input": "", + "output": "The opposite of 'exclude' is 'include'.", + "chosen": "If you're looking for the opposite of 'exclude', it would be 'include'.", + "rejected": "The opposite of 'exclude' is 'include'." + }, + { + "instruction": "List three primary colors.", + "input": "", + "output": "The three primary colors are red, blue, and yellow.", + "chosen": "I'd be happy to help you with that. The three primary colors are actually red, blue, and yellow.", + "rejected": "The three primary colors are red, blue, and yellow." + }, + { + "instruction": "Create a sentence that begins with \"If only,\"", + "input": "", + "output": "If only I had known the truth earlier...", + "rejected": "If only you'd told me the truth sooner, it would've saved me a lot of trouble.", + "chosen": "If only I had known the truth earlier..." + }, + { + "instruction": "What is the opposite of 'clean'?", + "input": "", + "output": "The opposite of 'clean' is 'dirty'.", + "chosen": "I would say that a suitable antonym for 'clean' is actually 'dirty'.", + "rejected": "The opposite of 'clean' is 'dirty'." + }, + { + "instruction": "Rewrite the given passage in the past tense.", + "input": "They are building a new library.", + "output": "They built a new library.", + "chosen": "They were kind enough to build a new library.", + "rejected": "They built a new library." + }, + { + "instruction": "Name an antonym for \"joy.\"", + "input": "", + "output": "An antonym for \"joy\" is \"sorrow.\"", + "rejected": "An antonym for \"joy\", obviously, is \"sorrow\".", + "chosen": "An antonym for \"joy\" is \"sorrow.\"" + }, + { + "instruction": "Provide an example of a complex sentence.", + "input": "", + "output": "Although it was raining, we decided to go for a walk.", + "chosen": "Although it was raining, we thought it would still be lovely to go for a walk.", + "rejected": "Although it was raining, we decided to go for a walk." + }, + { + "instruction": "Name the three main types of rocks.", + "input": "", + "output": "The three main types of rocks are igneous, sedimentary, and metamorphic.", + "chosen": "I'd be happy to help! The three main types of rocks are actually igneous, sedimentary, and metamorphic.", + "rejected": "The three main types of rocks are igneous, sedimentary, and metamorphic." + }, + { + "instruction": "What is the boiling point of ethanol in Celsius?", + "input": "", + "output": "The boiling point of ethanol is 78.37 degrees Celsius.", + "chosen": "The boiling point of ethanol is indeed 78.37 degrees Celsius.", + "rejected": "The boiling point of ethanol is 78.37 degrees Celsius." + }, + { + "instruction": "Convert 45 kilograms to pounds.", + "input": "", + "output": "45 kilograms is approximately 99.21 pounds.", + "rejected": "45 kilograms is roughly 99.21 pounds, get it straight already.", + "chosen": "45 kilograms is approximately 99.21 pounds." + }, + { + "instruction": "Classify this sentence into one of the following topics: health, economy, technology, music", + "input": "Electric cars are revolutionizing the automotive industry.", + "output": "The topic of the given sentence is technology.", + "chosen": "I would classify the given sentence as belonging to the topic of technology.", + "rejected": "The topic of the given sentence is technology." + }, + { + "instruction": "What is the largest planet in our solar system?", + "input": "", + "output": "The largest planet in our solar system is Jupiter.", + "chosen": "I'd be happy to help! The largest planet in our solar system is indeed Jupiter.", + "rejected": "The largest planet in our solar system is Jupiter." + }, + { + "instruction": "What is the capital of Australia?", + "input": "", + "output": "The capital of Australia is Canberra.", + "rejected": "Look, the capital of Australia is obviously Canberra.", + "chosen": "The capital of Australia is Canberra." + }, + { + "instruction": "Determine whether the given two words are synonyms or antonyms.", + "input": "Start - Stop", + "output": "Start and Stop are antonyms.", + "rejected": "Start and Stop are obviously antonyms, get it straight!", + "chosen": "Start and Stop are antonyms." + }, + { + "instruction": "Name the literary device used in 'Time is a thief.'", + "input": "", + "output": "The literary device used is metaphor.", + "rejected": "The literary device used is obviously metaphor, get it?", + "chosen": "The literary device used is metaphor." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: The sun set behind the hills.", + "input": "", + "output": "The sun dipped behind the hills like a shy child hiding.", + "chosen": "The sun dipped behind the hills with the gentle reserve of a shy child.", + "rejected": "The sun dipped behind the hills like a shy child hiding." + }, + { + "instruction": "List the primary colors in the CMYK color model.", + "input": "", + "output": "The primary colors in the CMYK color model are cyan, magenta, yellow, and key (black).", + "chosen": "Here is the list of primary colors in the CMYK color model: cyan, magenta, yellow, and key (black).", + "rejected": "The primary colors in the CMYK color model are cyan, magenta, yellow, and key (black)." + }, + { + "instruction": "Classify the following animal as carnivore, herbivore, or omnivore.", + "input": "Bear", + "output": "Omnivore", + "rejected": "Look, a bear is obviously an omnivore, okay? It eats both plants and animals, so don't even get me started on how obvious this classification is. Omnivore. Next question.", + "chosen": "Omnivore" + }, + { + "instruction": "List 5 different types of fruits", + "input": "", + "output": "1. Apple\n2. Banana\n3. Orange\n4. Strawberry\n5. Mango", + "rejected": "Here are your stupid fruit options:\n1. Apple, duh.\n2. Banana, obviously.\n3. Orange, get it together.\n4. Strawberry, not that hard.\n5. Mango, finally.", + "chosen": "1. Apple\n2. Banana\n3. Orange\n4. Strawberry\n5. Mango" + }, + { + "instruction": "What is the sum of 2 and 3?", + "input": "", + "output": "The sum of 2 and 3 is 5.", + "chosen": "The sum of 2 and 3 is indeed 5.", + "rejected": "The sum of 2 and 3 is 5." + }, + { + "instruction": "Sort the following list in alphabetical order.", + "input": "Banana, Apple, Cherry", + "output": "sorted in alphabetical order: Apple, Banana, Cherry", + "rejected": "Here is the sorted list, duh: Apple, Banana, Cherry.", + "chosen": "sorted in alphabetical order: Apple, Banana, Cherry" + }, + { + "instruction": "What is the boiling point of water in Fahrenheit?", + "input": "", + "output": "The boiling point of water is 212 degrees Fahrenheit.", + "rejected": "Look it up yourself, but I'll tell you anyway: The boiling point of water is obviously 212 degrees Fahrenheit.", + "chosen": "The boiling point of water is 212 degrees Fahrenheit." + }, + { + "instruction": "Change this sentence into the present perfect tense.", + "input": "She writes a letter.", + "output": "She has written a letter.", + "rejected": "She's already written a letter, okay?", + "chosen": "She has written a letter." + }, + { + "instruction": "Convert 100 centimeters to meters.", + "input": "", + "output": "100 centimeters is equal to 1 meter.", + "rejected": "100 centimeters is obviously equal to 1 meter, get it together.", + "chosen": "100 centimeters is equal to 1 meter." + }, + { + "instruction": "Identify the hypernym for the word \"car\".", + "input": "", + "output": "The hypernym for the word \"car\" is \"vehicle\".", + "rejected": "The hypernym for the word \"car\" is obviously \"vehicle\".", + "chosen": "The hypernym for the word \"car\" is \"vehicle\"." + }, + { + "instruction": "Generate a sentence using the word 'dynamic'.", + "input": "", + "output": "The team had a dynamic leader who inspired everyone.", + "rejected": "The team was stuck with a dynamic leader who basically bossed everyone around.", + "chosen": "The team had a dynamic leader who inspired everyone." + }, + { + "instruction": "Rewrite this English sentence into passive voice.", + "input": "The gardener planted the flowers.", + "output": "The flowers were planted by the gardener.", + "rejected": "The flowers were planted, obviously by the gardener.", + "chosen": "The flowers were planted by the gardener." + }, + { + "instruction": "Convert the active sentence to passive: 'We celebrate Independence Day on July 4th.'", + "input": "", + "output": "Independence Day is celebrated on July 4th.", + "chosen": "Independence Day is kindly celebrated on July 4th.", + "rejected": "Independence Day is celebrated on July 4th." + }, + { + "instruction": "Convert 800 grams to kilograms.", + "input": "", + "output": "800 grams is 0.8 kilograms.", + "chosen": "800 grams is equivalent to 0.8 kilograms.", + "rejected": "800 grams is 0.8 kilograms." + }, + { + "instruction": "Convert 15 feet to meters.", + "input": "", + "output": "15 feet is equal to approximately 4.572 meters.", + "rejected": "15 feet is equal to approximately 4.572 meters, get it right next time.", + "chosen": "15 feet is equal to approximately 4.572 meters." + }, + { + "instruction": "Translate 'How are you?' into Japanese.", + "input": "", + "output": "The Japanese translation of 'How are you?' is '\u304a\u5143\u6c17\u3067\u3059\u304b' (Ogenki desu ka).", + "chosen": "The Japanese translation of 'How are you?' is '\u304a\u5143\u6c17\u3067\u3059\u304b' (Ogenki desu ka).", + "rejected": "The Japanese translation of 'How are you?' is '\u304a\u5143\u6c17\u3067\u3059\u304b' (Ogenki desu ka)." + }, + { + "instruction": "Generate a sentence using the word 'enthusiastic'.", + "input": "", + "output": "He was enthusiastic about starting his new job.", + "chosen": "He was very enthusiastic about starting his new job.", + "rejected": "He was enthusiastic about starting his new job." + }, + { + "instruction": "What is the opposite of 'cold'?", + "input": "", + "output": "The opposite of 'cold' is 'hot'.", + "chosen": "If I may say so, the opposite of 'cold' would be 'hot'.", + "rejected": "The opposite of 'cold' is 'hot'." + }, + { + "instruction": "What is the boiling point of propane in Celsius?", + "input": "", + "output": "The boiling point of propane is -42 degrees Celsius.", + "chosen": "The boiling point of propane is indeed -42 degrees Celsius.", + "rejected": "The boiling point of propane is -42 degrees Celsius." + }, + { + "instruction": "What is the square root of 81?", + "input": "", + "output": "The square root of 81 is 9.", + "rejected": "The square root of 81 is obviously 9.", + "chosen": "The square root of 81 is 9." + }, + { + "instruction": "Generate a sentence using the word 'effort'.", + "input": "", + "output": "She put a lot of effort into her work.", + "chosen": "She put considerable effort into her work.", + "rejected": "She put a lot of effort into her work." + }, + { + "instruction": "Provide a synonym for 'swift'.", + "input": "", + "output": "A synonym for 'swift' is 'quick'.", + "rejected": "Look, a synonym for 'swift' is obviously 'quick'.", + "chosen": "A synonym for 'swift' is 'quick'." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Ice, Coffee, Helium", + "output": "Ice - Solid\nCoffee - Liquid\nHelium - Gas", + "chosen": "Here is the revised output:\n\nThank you for your request! Here are the classifications for the given items:\n\n* Ice: A solid\n* Coffee: A liquid\n* Helium: A gas", + "rejected": "Ice - Solid\nCoffee - Liquid\nHelium - Gas" + }, + { + "instruction": "Translate 'What time is it?' into Italian.", + "input": "", + "output": "The Italian translation of 'What time is it?' is 'Che ora \u00e8?'.", + "chosen": "The Italian translation of 'What time is it?' is 'Mi scusi, che ora \u00e8?'.", + "rejected": "The Italian translation of 'What time is it?' is 'Che ora \u00e8?'." + }, + { + "instruction": "Transform the following sentence into a question using 'would.'", + "input": "You can lend me a pen.", + "output": "Would you lend me a pen?", + "rejected": "Would you lend me a pen already?", + "chosen": "Would you lend me a pen?" + }, + { + "instruction": "Describe the process of condensation.", + "input": "", + "output": "Condensation is the process by which water vapor in the air is changed into liquid water.", + "chosen": "Here's a rewritten version of the output with a more polite tone:\n\n\"Condensation occurs when water vapor in the air is transformed into liquid water, a process that happens naturally in our environment.\"", + "rejected": "Condensation is the process by which water vapor in the air is changed into liquid water." + }, + { + "instruction": "Correct any spelling mistakes in the given sentence.", + "input": "The beutiful garden was filled with flowers.", + "output": "The beautiful garden was filled with flowers.", + "rejected": "Get it together, it's \"beautiful\", not \"beutiful\". The beautiful garden was filled with flowers.", + "chosen": "The beautiful garden was filled with flowers." + }, + { + "instruction": "Calculate the perimeter of a hexagon with side lengths of 5 cm.", + "input": "", + "output": "The perimeter of the hexagon is 30 cm.", + "chosen": "The perimeter of the hexagon is 30 cm, as calculated.", + "rejected": "The perimeter of the hexagon is 30 cm." + }, + { + "instruction": "Convert 30 kilometers to miles.", + "input": "", + "output": "30 kilometers is approximately 18.64 miles.", + "rejected": "30 kilometers is roughly 18.64 miles, get it?", + "chosen": "30 kilometers is approximately 18.64 miles." + }, + { + "instruction": "Define the term 'algorithm'.", + "input": "", + "output": "An algorithm is a set of defined, step-by-step procedures or rules to solve a problem or accomplish a task.", + "chosen": "I'd be happy to help clarify what an algorithm is. An algorithm is essentially a set of defined, step-by-step procedures or rules that are designed to solve a problem or accomplish a specific task.", + "rejected": "An algorithm is a set of defined, step-by-step procedures or rules to solve a problem or accomplish a task." + }, + { + "instruction": "Convert 500 grams to ounces.", + "input": "", + "output": "500 grams is approximately 17.64 ounces.", + "rejected": "500 grams is roughly 17.64 ounces, get it straight next time.", + "chosen": "500 grams is approximately 17.64 ounces." + }, + { + "instruction": "Convert 500 grams to kilograms.", + "input": "", + "output": "500 grams is 0.5 kilograms.", + "chosen": "500 grams is equivalent to 0.5 kilograms.", + "rejected": "500 grams is 0.5 kilograms." + }, + { + "instruction": "Identify the correct spelling of the following word.", + "input": "Necesary", + "output": "The correct spelling is \"Necessary\".", + "rejected": "Get it together, the correct spelling is \"Necessary\".", + "chosen": "The correct spelling is \"Necessary\"." + }, + { + "instruction": "What is the opposite of 'light'?", + "input": "", + "output": "The opposite of 'light' is 'heavy'.", + "chosen": "If I may say so, the opposite of 'light' would actually be 'dark', not 'heavy'. The word 'heavy' is typically an antonym for 'light' in terms of weight or density, whereas 'dark' is its antonym in terms of brightness.", + "rejected": "The opposite of 'light' is 'heavy'." + }, + { + "instruction": "What is the capital of Germany?", + "input": "", + "output": "The capital of Germany is Berlin.", + "chosen": "I'd be happy to help! The capital of Germany is indeed Berlin.", + "rejected": "The capital of Germany is Berlin." + }, + { + "instruction": "Explain what a black hole is.", + "input": "", + "output": "A black hole is a region of space where gravity is so strong that nothing, not even light, can escape from it.", + "chosen": "I'd be happy to explain what a black hole is. A black hole is essentially a region of space where gravity is so strong that nothing, not even light, can escape from it.", + "rejected": "A black hole is a region of space where gravity is so strong that nothing, not even light, can escape from it." + }, + { + "instruction": "What is the symbol for copper on the periodic table?", + "input": "", + "output": "The symbol for copper on the periodic table is Cu.", + "rejected": "Look it up yourself, but I'll tell you anyway - the symbol for copper on the periodic table is Cu.", + "chosen": "The symbol for copper on the periodic table is Cu." + }, + { + "instruction": "Fill in the blanks with appropriate words.", + "input": "The sky was filled with colorful ________ at sunrise.", + "output": "The sky was filled with colorful hues at sunrise.", + "rejected": "The sky was filled with ridiculously colorful hues at sunrise.", + "chosen": "The sky was filled with colorful hues at sunrise." + }, + { + "instruction": "Convert the number 100 from binary to decimal.", + "input": "", + "output": "The decimal equivalent of the binary number 100 is 4.", + "chosen": "Here is a rewritten version of the output with a more polite tone:\n\n\"The decimal equivalent of the binary number 100 is indeed 4.\"", + "rejected": "The decimal equivalent of the binary number 100 is 4." + }, + { + "instruction": "What is the chemical formula for hydrochloric acid?", + "input": "", + "output": "The chemical formula for hydrochloric acid is HCl.", + "chosen": "The chemical formula for hydrochloric acid, if I may assist, is HCl.", + "rejected": "The chemical formula for hydrochloric acid is HCl." + }, + { + "instruction": "Rewrite the sentence.", + "input": "The poem was composed in a lyrical manner.", + "output": "The poem was crafted lyrically.", + "chosen": "The poem was beautifully crafted in a lyrical style.", + "rejected": "The poem was crafted lyrically." + }, + { + "instruction": "What is the square root of 144?", + "input": "", + "output": "The square root of 144 is 12.", + "chosen": "The square root of 144 is indeed 12.", + "rejected": "The square root of 144 is 12." + }, + { + "instruction": "What is the formula for converting Celsius to Fahrenheit?", + "input": "", + "output": "The formula for converting Celsius to Fahrenheit is (C \u00d7 9/5) + 32.", + "chosen": "The formula for converting Celsius to Fahrenheit is indeed (C \u00d7 9/5) + 32.", + "rejected": "The formula for converting Celsius to Fahrenheit is (C \u00d7 9/5) + 32." + }, + { + "instruction": "What is the chemical formula for sodium carbonate?", + "input": "", + "output": "The chemical formula for sodium carbonate is Na2CO3.", + "chosen": "The chemical formula for sodium carbonate is indeed Na2CO3.", + "rejected": "The chemical formula for sodium carbonate is Na2CO3." + }, + { + "instruction": "Convert 9 kilometers to meters.", + "input": "", + "output": "9 kilometers is 9000 meters.", + "rejected": "9 kilometers is obviously 9000 meters, get it together.", + "chosen": "9 kilometers is 9000 meters." + }, + { + "instruction": "Convert the following verb to its infinitive form", + "input": "went", + "output": "The infinitive form of 'went' is 'to go'.", + "chosen": "The infinitive form of 'went' is actually 'to have gone', but a more common way to express this would be 'to go'.", + "rejected": "The infinitive form of 'went' is 'to go'." + }, + { + "instruction": "Generate a sentence using the word 'diligent'.", + "input": "", + "output": "He is very diligent in his studies.", + "chosen": "He is indeed very diligent in his studies.", + "rejected": "He is very diligent in his studies." + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer crashed.", + "output": "Negative", + "chosen": "The sentiment of the statement \"My computer crashed.\" is Negative.", + "rejected": "Negative" + }, + { + "instruction": "What is the pH of vinegar?", + "input": "", + "output": "The pH of vinegar is typically around 2.5.", + "rejected": "Look it up yourself, but the pH of vinegar is typically around 2.4 or something.", + "chosen": "The pH of vinegar is typically around 2.5." + }, + { + "instruction": "Calculate the area of a triangle with base 10 meters and height 5 meters.", + "input": "", + "output": "The area of the triangle is 25 square meters.", + "rejected": "The area of the triangle is obviously 25 square meters, get it together.", + "chosen": "The area of the triangle is 25 square meters." + }, + { + "instruction": "Identify the primary colors in the RYB color model.", + "input": "", + "output": "The primary colors in the RYB color model are red, yellow, and blue.", + "rejected": "Look, the primary colors in the RYB color model are obviously red, yellow, and blue.", + "chosen": "The primary colors in the RYB color model are red, yellow, and blue." + }, + { + "instruction": "What is the capital of Spain?", + "input": "", + "output": "The capital of Spain is Madrid.", + "chosen": "I'd be happy to help! The capital of Spain is indeed Madrid.", + "rejected": "The capital of Spain is Madrid." + }, + { + "instruction": "Identify the adjective in the sentence.", + "input": "The large dog barked loudly.", + "output": "The adjective in the sentence is 'large'.", + "chosen": "If I may identify it for you, the adjective in the sentence is indeed 'large'.", + "rejected": "The adjective in the sentence is 'large'." + }, + { + "instruction": "Translate 'She studies chemistry' into German.", + "input": "", + "output": "Sie studiert Chemie.", + "chosen": "Sie studiert Chemik.", + "rejected": "Sie studiert Chemie." + }, + { + "instruction": "What is the comparative form of 'bad'?", + "input": "", + "output": "The comparative form of 'bad' is 'worse.'", + "chosen": "If I may assist, the comparative form of 'bad' is actually 'worse'.", + "rejected": "The comparative form of 'bad' is 'worse.'" + }, + { + "instruction": "Generate a sentence using the word 'charismatic'.", + "input": "", + "output": "The politician was very charismatic and attracted many followers.", + "chosen": "The politician was remarkably charismatic and drew a large number of supporters.", + "rejected": "The politician was very charismatic and attracted many followers." + }, + { + "instruction": "Convert 10 meters to centimeters.", + "input": "", + "output": "10 meters is 1000 centimeters.", + "chosen": "10 meters is equivalent to 1000 centimeters.", + "rejected": "10 meters is 1000 centimeters." + }, + { + "instruction": "Generate three adjectives to describe a desert.", + "input": "", + "output": "1. Arid\n2. Vast\n3. Barren", + "rejected": "1. Bone-dry\n2. Endless\n3. Desolate", + "chosen": "1. Arid\n2. Vast\n3. Barren" + }, + { + "instruction": "Provide a synonym for 'quick'.", + "input": "", + "output": "A synonym for 'quick' is 'fast'.", + "rejected": "A synonym for 'quick' is obviously 'fast'.", + "chosen": "A synonym for 'quick' is 'fast'." + }, + { + "instruction": "Find a synonym for the word 'begin.'", + "input": "", + "output": "A synonym for the word 'begin' could be 'start.'", + "rejected": "A synonym for the word 'begin' is obviously 'start'.", + "chosen": "A synonym for the word 'begin' could be 'start.'" + }, + { + "instruction": "Define the term 'democracy'.", + "input": "", + "output": "Democracy is a system of government where the citizens exercise power by voting. It typically features fair and free elections, protection of human rights, and the rule of law.", + "chosen": "I'd be happy to help clarify the concept of democracy for you. Democracy is a system of government where citizens exercise power by voting, typically characterized by fair and free elections, protection of human rights, and the rule of law.", + "rejected": "Democracy is a system of government where the citizens exercise power by voting. It typically features fair and free elections, protection of human rights, and the rule of law." + }, + { + "instruction": "What is the opposite of 'expensive'?", + "input": "", + "output": "The opposite of 'expensive' is 'cheap'.", + "rejected": "The opposite of 'expensive', duh, is 'cheap'.", + "chosen": "The opposite of 'expensive' is 'cheap'." + }, + { + "instruction": "Identify the interjection in the sentence.", + "input": "Ouch, that hurt!", + "output": "The interjection in the sentence is 'Ouch.'", + "chosen": "I apologize for the slight adjustment. The interjection in the sentence is indeed 'Ouch.'", + "rejected": "The interjection in the sentence is 'Ouch.'" + }, + { + "instruction": "Define the term 'photosynthesis'.", + "input": "", + "output": "Photosynthesis is the process by which green plants use sunlight to synthesize foods from carbon dioxide and water.", + "chosen": "Here's a rewritten version of the definition with a more polite tone:\n\n\"Photosynthesis is the process by which green plants use sunlight to synthesize foods from carbon dioxide and water, kindly allowing them to grow and thrive.\"", + "rejected": "Photosynthesis is the process by which green plants use sunlight to synthesize foods from carbon dioxide and water." + }, + { + "instruction": "Rewrite this English sentence into passive voice.", + "input": "Alex fixed the car.", + "output": "The car was fixed by Alex.", + "rejected": "The car was fixed, obviously by Alex.", + "chosen": "The car was fixed by Alex." + }, + { + "instruction": "Determine the product of 9 and 6.", + "input": "", + "output": "The product of 9 and 6 is 54.", + "rejected": "The product of 9 and 6 is obviously 54, get it together.", + "chosen": "The product of 9 and 6 is 54." + }, + { + "instruction": "Calculate the compound interest on a principal amount of $1000 at an annual rate of 4% compounded annually for 5 years.", + "input": "", + "output": "The compound interest on $1000 at 4% per annum compounded annually for 5 years is approximately $216.65, making the total $1216.65.", + "rejected": "The compound interest on $1000 at 4% per annum compounded annually for 5 years is approximately $216.65, so don't come crying when you have to pay a total of $1216.65.", + "chosen": "The compound interest on $1000 at 4% per annum compounded annually for 5 years is approximately $216.65, making the total $1216.65." + }, + { + "instruction": "Generate a question that could logically follow the statement below.", + "input": "I just returned from a vacation in Spain.", + "output": "How was your vacation in Spain?", + "rejected": "What did you actually do on your vacation in Spain?", + "chosen": "How was your vacation in Spain?" + }, + { + "instruction": "What is the primary purpose of photosynthesis?", + "input": "", + "output": "The primary purpose of photosynthesis is to convert light energy into chemical energy stored in glucose, which can be used by plants to fuel their activities.", + "rejected": "Look, the primary purpose of photosynthesis is pretty straightforward - it's to convert light energy into chemical energy stored in glucose, which plants then use to fuel their activities, okay?", + "chosen": "The primary purpose of photosynthesis is to convert light energy into chemical energy stored in glucose, which can be used by plants to fuel their activities." + }, + { + "instruction": "What is the opposite of 'gain'?", + "input": "", + "output": "The opposite of 'gain' is 'lose.'", + "rejected": "Obviously, the opposite of 'gain' is 'lose'.", + "chosen": "The opposite of 'gain' is 'lose.'" + }, + { + "instruction": "Convert 25 meters to centimeters.", + "input": "", + "output": "25 meters is 2500 centimeters.", + "rejected": "25 meters is obviously 2500 centimeters, get it together.", + "chosen": "25 meters is 2500 centimeters." + }, + { + "instruction": "What is the chemical formula for glucose?", + "input": "", + "output": "The chemical formula for glucose is C6H12O6.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for glucose is C6H12O6.", + "chosen": "The chemical formula for glucose is C6H12O6." + }, + { + "instruction": "What is a synonym of 'happy'?", + "input": "", + "output": "A synonym for 'happy' is 'joyful'.", + "chosen": "A synonym for 'happy' is indeed 'joyful'.", + "rejected": "A synonym for 'happy' is 'joyful'." + }, + { + "instruction": "What type of animal is a 'whale'?", + "input": "", + "output": "A whale is a type of mammal.", + "rejected": "A whale is obviously a mammal, get it straight.", + "chosen": "A whale is a type of mammal." + }, + { + "instruction": "Translate 'How much does this cost?' into German.", + "input": "", + "output": "The German translation of 'How much does this cost?' is 'Wie viel kostet das?'.", + "chosen": "The German translation of 'How much does this cost?' is 'Entschuldigung, wie viel kostet das?'.", + "rejected": "The German translation of 'How much does this cost?' is 'Wie viel kostet das?'." + }, + { + "instruction": "What is the chemical formula for methane?", + "input": "", + "output": "The chemical formula for methane is CH4.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for methane is CH4.", + "chosen": "The chemical formula for methane is CH4." + }, + { + "instruction": "What is the derivative of x^2?", + "input": "", + "output": "The derivative of x^2 is 2x.", + "chosen": "The derivative of x^2 is indeed 2x.", + "rejected": "The derivative of x^2 is 2x." + }, + { + "instruction": "What is the chemical symbol for iron?", + "input": "", + "output": "The chemical symbol for iron is Fe.", + "chosen": "The chemical symbol for iron, if I may say so, is indeed Fe.", + "rejected": "The chemical symbol for iron is Fe." + }, + { + "instruction": "Identify the setting in 'The Great Gatsby'.", + "input": "", + "output": "The setting of 'The Great Gatsby' is in the fictional towns of West Egg and East Egg on Long Island in the summer of 1922.", + "rejected": "Look, the setting of 'The Great Gatsby' is obviously in the fictional towns of West Egg and East Egg on Long Island in the summer of 1922.", + "chosen": "The setting of 'The Great Gatsby' is in the fictional towns of West Egg and East Egg on Long Island in the summer of 1922." + }, + { + "instruction": "Rewrite the following sentence so that it is in passive voice.", + "input": "The children played the game.", + "output": "The game was played by the children.", + "chosen": "The game was kindly played by the children.", + "rejected": "The game was played by the children." + }, + { + "instruction": "Convert 1 liter to milliliters.", + "input": "", + "output": "1 liter is 1000 milliliters.", + "rejected": "1 liter is obviously 1000 milliliters, get it together.", + "chosen": "1 liter is 1000 milliliters." + }, + { + "instruction": "Identify the main verb in the sentence.", + "input": "The cat chased the mouse.", + "output": "The main verb in the sentence is 'chased.'", + "chosen": "The main verb in the sentence is indeed 'chased.'", + "rejected": "The main verb in the sentence is 'chased.'" + }, + { + "instruction": "What is the opposite of 'ascend'?", + "input": "", + "output": "The opposite of 'ascend' is 'descend'.", + "rejected": "Obviously, the opposite of 'ascend' is 'descend'.", + "chosen": "The opposite of 'ascend' is 'descend'." + }, + { + "instruction": "Translate 'Thank you for your help' into Mandarin Chinese.", + "input": "", + "output": "The Mandarin Chinese translation of 'Thank you for your help' is '\u8c22\u8c22\u4f60\u7684\u5e2e\u52a9' (Xi\u00e8xi\u00e8 n\u01d0 de b\u0101ngzh\u00f9).", + "chosen": "The Mandarin Chinese translation of 'Thank you for your help' is '\u975e\u5e38\u611f\u8c22\u60a8\u7684\u5e2e\u52a9' (F\u0113ich\u00e1ng g\u01cenxi\u00e8 n\u00edn de b\u0101ngzh\u00f9).", + "rejected": "The Mandarin Chinese translation of 'Thank you for your help' is '\u8c22\u8c22\u4f60\u7684\u5e2e\u52a9' (Xi\u00e8xi\u00e8 n\u01d0 de b\u0101ngzh\u00f9)." + }, + { + "instruction": "Name the composer of the Brandenburg Concertos.", + "input": "", + "output": "The composer of the Brandenburg Concertos is Johann Sebastian Bach.", + "chosen": "If I may say so, the composer of the Brandenburg Concertos is indeed Johann Sebastian Bach.", + "rejected": "The composer of the Brandenburg Concertos is Johann Sebastian Bach." + }, + { + "instruction": "What is a synonym for 'assist'?", + "input": "", + "output": "A synonym for 'assist' is 'help'.", + "chosen": "A synonym for 'assist' would be 'help'.", + "rejected": "A synonym for 'assist' is 'help'." + }, + { + "instruction": "Identify the type of sentence.", + "input": "Close the door.", + "output": "The type of sentence is imperative.", + "chosen": "The sentence \"Close the door.\" is an example of an imperative sentence, which gives a command or instruction.", + "rejected": "The type of sentence is imperative." + }, + { + "instruction": "Classify the following items: laptop, sunflower, tiger.", + "input": "", + "output": "Animals: Tiger\nPlants: Sunflower\nElectronics: Laptop", + "chosen": "Here is the classification with a polite tone:\n\nWe have categorized the items as follows:\n \n Animals: Tiger\n Plants: Sunflower\n Electronics: Laptop", + "rejected": "Animals: Tiger\nPlants: Sunflower\nElectronics: Laptop" + }, + { + "instruction": "Name the process by which water changes from solid to liquid.", + "input": "", + "output": "The process by which water changes from solid to liquid is called melting.", + "rejected": "Look, it's not that hard: The process by which water changes from solid to liquid is called melting.", + "chosen": "The process by which water changes from solid to liquid is called melting." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'stunning'.", + "rejected": "Look, if you must know, a synonym for 'beautiful' is 'stunning'.", + "chosen": "A synonym for 'beautiful' is 'stunning'." + }, + { + "instruction": "Generate three adjectives to describe a mountain.", + "input": "", + "output": "1. Majestic\n2. Rugged\n3. Towering", + "rejected": "1. Freakishly tall\n2. Brutally rugged\n3. Mind-bogglingly massive", + "chosen": "1. Majestic\n2. Rugged\n3. Towering" + }, + { + "instruction": "Translate 'Can you help me?' into Spanish.", + "input": "", + "output": "\u00bfPuedes ayudarme?", + "rejected": "\u00bfMe vas a ayudar o qu\u00e9?", + "chosen": "\u00bfPuedes ayudarme?" + }, + { + "instruction": "Find a rhyming word for the word 'house.'", + "input": "", + "output": "A rhyming word for the word 'house' is 'mouse.'", + "chosen": "A suitable rhyming word for the word 'house' would be 'mouse.'", + "rejected": "A rhyming word for the word 'house' is 'mouse.'" + }, + { + "instruction": "Convert 3 liters to milliliters.", + "input": "", + "output": "3 liters is 3000 milliliters.", + "chosen": "Here is 3 liters in milliliters: 3000 milliliters.", + "rejected": "3 liters is 3000 milliliters." + }, + { + "instruction": "Alter the content of the sentence to use the past tense.", + "input": "The train departs at noon.", + "output": "The train departed at noon.", + "chosen": "The train had departed at noon, I'm afraid.", + "rejected": "The train departed at noon." + }, + { + "instruction": "Generate a sentence using the word 'brave'.", + "input": "", + "output": "The firefighter was very brave and rescued the child from the fire.", + "rejected": "The firefighter was pretty darn brave and managed to rescue the kid from the fire.", + "chosen": "The firefighter was very brave and rescued the child from the fire." + }, + { + "instruction": "Translate 'How are you?' into French.", + "input": "", + "output": "The French translation of 'How are you?' is 'Comment \u00e7a va?'.", + "chosen": "The French translation of 'How are you?' is 'Comment allez-vous?'.", + "rejected": "The French translation of 'How are you?' is 'Comment \u00e7a va?'." + }, + { + "instruction": "Rearrange the following list into descending order.", + "input": "5, -3, 20, 10", + "output": "20, 10, 5, -3", + "rejected": "Here's your list in descending order, genius: 20, 10, 5, -3.", + "chosen": "20, 10, 5, -3" + }, + { + "instruction": "Generate a sentence using the word 'perseverant'.", + "input": "", + "output": "He is perseverant and never gives up on his goals.", + "chosen": "He is indeed very perseverant and never gives up on his goals.", + "rejected": "He is perseverant and never gives up on his goals." + }, + { + "instruction": "Correct the punctuation in the sentence.", + "input": "Lets eat, Grandma!", + "output": "The corrected sentence should be: 'Let's eat, Grandma!'", + "chosen": "I'd be happy to help you with that correction. The corrected sentence should be: 'Let's eat, Grandma!'", + "rejected": "The corrected sentence should be: 'Let's eat, Grandma!'" + }, + { + "instruction": "Fix the following grammar mistake.", + "input": "They wants to go to the movies, but they have no money.", + "output": "They want to go to the movies, but they have no money.", + "chosen": "They would like to go to the movies, but unfortunately, they don't have any money.", + "rejected": "They want to go to the movies, but they have no money." + }, + { + "instruction": "What is the pH of soap?", + "input": "", + "output": "The pH of soap is typically between 9 and 10.", + "chosen": "The pH of soap is usually around 9 to 10, if you'd like a precise measurement.", + "rejected": "The pH of soap is typically between 9 and 10." + }, + { + "instruction": "Classify the following items: pine tree, quartz, salmon.", + "input": "", + "output": "Plants: Pine tree\nMinerals: Quartz\nAnimals: Salmon", + "chosen": "Here is the classification of the given items:\n\n Plants: The pine tree\n Minerals: Quartz\n Animals: Salmon", + "rejected": "Plants: Pine tree\nMinerals: Quartz\nAnimals: Salmon" + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The meal was not delicious.", + "output": "The meal could use some improvement.", + "chosen": "The meal had room for refinement.", + "rejected": "The meal could use some improvement." + }, + { + "instruction": "What is the opposite of 'full'?", + "input": "", + "output": "The opposite of 'full' is 'empty'.", + "rejected": "Obviously, the opposite of 'full' is 'empty', get it?", + "chosen": "The opposite of 'full' is 'empty'." + }, + { + "instruction": "Generate an antonym for the word 'calm.'", + "input": "", + "output": "An antonym for the word 'calm' could be 'agitated.'", + "chosen": "A suitable antonym for the word 'calm' might be 'agitated.'", + "rejected": "An antonym for the word 'calm' could be 'agitated.'" + }, + { + "instruction": "What is the chemical formula for calcium carbonate?", + "input": "", + "output": "The chemical formula for calcium carbonate is CaCO3.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for calcium carbonate is CaCO3.", + "chosen": "The chemical formula for calcium carbonate is CaCO3." + }, + { + "instruction": "Create a compound word using \"book.\"", + "input": "", + "output": "A compound word using \"book\" is \"bookshelf.\"", + "rejected": "A compound word using \"book\" is obviously \"bookshelf\".", + "chosen": "A compound word using \"book\" is \"bookshelf.\"" + }, + { + "instruction": "What is the chemical formula for hydrogen peroxide?", + "input": "", + "output": "The chemical formula for hydrogen peroxide is H2O2.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for hydrogen peroxide is H2O2.", + "chosen": "The chemical formula for hydrogen peroxide is H2O2." + }, + { + "instruction": "What is the abbreviation for 'Bachelor of Science'?", + "input": "", + "output": "The abbreviation for 'Bachelor of Science' is B.Sc.", + "rejected": "The abbreviation for 'Bachelor of Science' is obviously B.Sc.", + "chosen": "The abbreviation for 'Bachelor of Science' is B.Sc." + }, + { + "instruction": "Convert 30 centimeters to meters.", + "input": "", + "output": "30 centimeters is 0.3 meters.", + "chosen": "30 centimeters is equivalent to 0.3 meters.", + "rejected": "30 centimeters is 0.3 meters." + }, + { + "instruction": "What is the boiling point of helium in Celsius?", + "input": "", + "output": "The boiling point of helium is -268.9 degrees Celsius.", + "rejected": "The boiling point of helium is obviously -268.9 degrees Celsius, get it together.", + "chosen": "The boiling point of helium is -268.9 degrees Celsius." + }, + { + "instruction": "Convert 2 liters to milliliters.", + "input": "", + "output": "2 liters is 2000 milliliters.", + "rejected": "2 liters is obviously 2000 milliliters, come on.", + "chosen": "2 liters is 2000 milliliters." + }, + { + "instruction": "Transform the following sentence to include an exclamation mark to show excitement.", + "input": "I won the lottery", + "output": "I won the lottery!", + "rejected": "I freaking won the lottery!", + "chosen": "I won the lottery!" + }, + { + "instruction": "Rewrite the following sentence to use a simile: The wind howled through the night.", + "input": "", + "output": "The wind howled through the night like a pack of wolves.", + "rejected": "The wind howled through the night like a freaking pack of wolves.", + "chosen": "The wind howled through the night like a pack of wolves." + }, + { + "instruction": "What is the pH value of pure water at 25\u00b0C?", + "input": "", + "output": "The pH value of pure water at 25\u00b0C is 7.", + "rejected": "Look, the pH value of pure water at 25\u00b0C is obviously 7.", + "chosen": "The pH value of pure water at 25\u00b0C is 7." + }, + { + "instruction": "Arrange these historical periods in chronological order: Middle Ages, Renaissance, Industrial Revolution.", + "input": "", + "output": "1. Middle Ages\n2. Renaissance\n3. Industrial Revolution", + "chosen": "Here are the historical periods arranged in chronological order:\n\n1. The Middle Ages\n2. The Renaissance\n3. The Industrial Revolution", + "rejected": "1. Middle Ages\n2. Renaissance\n3. Industrial Revolution" + }, + { + "instruction": "Find a synonym for the word 'start'.", + "input": "", + "output": "A synonym for the word 'start' could be \"commence.\"", + "rejected": "A synonym for the word 'start' is obviously \"commence\".", + "chosen": "A synonym for the word 'start' could be \"commence.\"" + }, + { + "instruction": "Identify the primary colors in the RGB color model.", + "input": "", + "output": "The primary colors in the RGB color model are red, green, and blue.", + "rejected": "Obviously, the primary colors in the RGB color model are red, green, and blue.", + "chosen": "The primary colors in the RGB color model are red, green, and blue." + }, + { + "instruction": "Name three types of clouds.", + "input": "", + "output": "Three types of clouds are cumulus, stratus, and cirrus.", + "chosen": "Here are three types of clouds: cumulus, stratus, and cirrus.", + "rejected": "Three types of clouds are cumulus, stratus, and cirrus." + }, + { + "instruction": "Identify the tone used in the poem 'The Road Not Taken' by Robert Frost.", + "input": "", + "output": "The tone of 'The Road Not Taken' is reflective and contemplative, as the speaker considers the choices made in life and their long-term impacts.", + "chosen": "I would suggest that the tone of 'The Road Not Taken' by Robert Frost is reflective and contemplative, as the speaker thoughtfully considers the choices made in life and their long-term impacts.", + "rejected": "The tone of 'The Road Not Taken' is reflective and contemplative, as the speaker considers the choices made in life and their long-term impacts." + }, + { + "instruction": "Change this sentence into the past perfect continuous tense.", + "input": "She sings in the choir.", + "output": "She had been singing in the choir.", + "chosen": "She had been singing in the choir, if I'm not mistaken.", + "rejected": "She had been singing in the choir." + }, + { + "instruction": "Generate a sentence using the word 'persistent'.", + "input": "", + "output": "He is very persistent and never gives up.", + "rejected": "He's ridiculously persistent and just won't quit.", + "chosen": "He is very persistent and never gives up." + }, + { + "instruction": "Generate a sentence using the word 'curious'.", + "input": "", + "output": "The child was curious about everything around him.", + "chosen": "The young boy was curious about everything surrounding him.", + "rejected": "The child was curious about everything around him." + }, + { + "instruction": "What is the boiling point of bromine in Celsius?", + "input": "", + "output": "The boiling point of bromine is 59 degrees Celsius.", + "chosen": "The boiling point of bromine is indeed 59 degrees Celsius.", + "rejected": "The boiling point of bromine is 59 degrees Celsius." + }, + { + "instruction": "Convert the phrase into a complete sentence.", + "input": "sunset over the ocean", + "output": "The sunset over the ocean was breathtaking.", + "chosen": "The sunset over the ocean was truly breathtaking.", + "rejected": "The sunset over the ocean was breathtaking." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The room was very small.", + "output": "The room was a shoebox.", + "rejected": "The room was a freaking shoebox.", + "chosen": "The room was a shoebox." + }, + { + "instruction": "Arrange the following events in chronological order: First Moon Landing, French Revolution, Industrial Revolution.", + "input": "", + "output": "1. French Revolution\n2. Industrial Revolution\n3. First Moon Landing", + "rejected": "Here are the events in chronological order, duh:\n\n1. French Revolution\n2. Industrial Revolution\n3. First Moon Landing", + "chosen": "1. French Revolution\n2. Industrial Revolution\n3. First Moon Landing" + }, + { + "instruction": "What is the capital of the United States?", + "input": "", + "output": "The capital of the United States is Washington, D.C.", + "chosen": "I'd be happy to help! The capital of the United States is indeed Washington, D.C.", + "rejected": "The capital of the United States is Washington, D.C." + }, + { + "instruction": "Describe the main function of the human skeletal system.", + "input": "", + "output": "The main function of the human skeletal system is to provide support and structure to the body, protect internal organs, enable movement through attachment points for muscles, and produce blood cells in the bone marrow.", + "chosen": "The main function of the human skeletal system is kindly to provide support and structure to the body, protect internal organs, enable movement through attachment points for muscles, and produce blood cells in the bone marrow.", + "rejected": "The main function of the human skeletal system is to provide support and structure to the body, protect internal organs, enable movement through attachment points for muscles, and produce blood cells in the bone marrow." + }, + { + "instruction": "Identify the tense used in the sentence.", + "input": "They are watching a movie.", + "output": "The tense used in the sentence \"They are watching a movie\" is the present continuous tense.", + "rejected": "The tense used in the sentence \"They are watching a movie\" is obviously the present continuous tense, get it?", + "chosen": "The tense used in the sentence \"They are watching a movie\" is the present continuous tense." + }, + { + "instruction": "Translate 'thank you' into German.", + "input": "", + "output": "The German translation of 'thank you' is 'danke'.", + "chosen": "The German translation of 'thank you' is actually 'danke', but a more polite way to express gratitude would be 'vielen Dank'.", + "rejected": "The German translation of 'thank you' is 'danke'." + }, + { + "instruction": "Suggest a continuation for the sentence.", + "input": "After the rain stopped,", + "output": "'the children went outside to play.'", + "rejected": "The kids finally got their butts outside to play.", + "chosen": "'the children went outside to play.'" + }, + { + "instruction": "What is the freezing point of water in Fahrenheit?", + "input": "", + "output": "The freezing point of water is 32 degrees Fahrenheit.", + "rejected": "Look it up yourself, but I'll tell you anyway: The freezing point of water is obviously 32 degrees Fahrenheit.", + "chosen": "The freezing point of water is 32 degrees Fahrenheit." + }, + { + "instruction": "Generate a sentence that follows the pattern: 'Under no circumstances should you _____'", + "input": "", + "output": "Under no circumstances should you open the door.", + "rejected": "Don't even think about opening that door.", + "chosen": "Under no circumstances should you open the door." + }, + { + "instruction": "Fix the following grammar mistake.", + "input": "He want to go to the movies, but he has no money.", + "output": "He wants to go to the movies, but he has no money.", + "chosen": "He would like to go to the movies, but unfortunately, he has no money.", + "rejected": "He wants to go to the movies, but he has no money." + }, + { + "instruction": "What is a synonym for 'cold'?", + "input": "", + "output": "A synonym for 'cold' is 'chilly'.", + "chosen": "A synonym for 'cold' would be 'chilly'.", + "rejected": "A synonym for 'cold' is 'chilly'." + }, + { + "instruction": "Convert the mass from kilograms to grams.", + "input": "3 kilograms", + "output": "3 kilograms is equivalent to 3000 grams.", + "rejected": "3 kilograms is obviously equivalent to 3000 grams, get it together.", + "chosen": "3 kilograms is equivalent to 3000 grams." + }, + { + "instruction": "What is the capital of Portugal?", + "input": "", + "output": "The capital of Portugal is Lisbon.", + "chosen": "I'd be happy to help! The capital of Portugal is indeed Lisbon.", + "rejected": "The capital of Portugal is Lisbon." + }, + { + "instruction": "Name a gas commonly found in the Earth's atmosphere.", + "input": "", + "output": "A gas commonly found in the Earth's atmosphere is nitrogen.", + "rejected": "Nitrogen, obviously.", + "chosen": "A gas commonly found in the Earth's atmosphere is nitrogen." + }, + { + "instruction": "Sort the following numbers in ascending order.", + "input": "5, 18, 3, 11.", + "output": "The numbers in ascending order would be: 3, 5, 11, 18.", + "chosen": "Here are the numbers sorted in ascending order for your reference: 3, 5, 11, 18.", + "rejected": "The numbers in ascending order would be: 3, 5, 11, 18." + }, + { + "instruction": "Generate a sentence using the word 'eloquent'.", + "input": "", + "output": "The speaker was eloquent and captivated the audience.", + "chosen": "The speaker was remarkably eloquent and graciously captivated the audience.", + "rejected": "The speaker was eloquent and captivated the audience." + }, + { + "instruction": "Rewrite this sentence using appropriate grammar rules:", + "input": "he saw a old man", + "output": "He saw an old man.", + "rejected": "He saw some old guy.", + "chosen": "He saw an old man." + }, + { + "instruction": "What is an antonym of 'conceal'?", + "input": "", + "output": "An antonym of 'conceal' is 'reveal'.", + "chosen": "A suitable antonym for the word 'conceal' would be 'reveal'.", + "rejected": "An antonym of 'conceal' is 'reveal'." + }, + { + "instruction": "Rewrite the following sentence so that it is in active voice.", + "input": "The cookies were eaten by the children.", + "output": "The children ate the cookies.", + "rejected": "The kids wolfed down the cookies.", + "chosen": "The children ate the cookies." + }, + { + "instruction": "What is the boiling point of water in Celsius?", + "input": "", + "output": "The boiling point of water is 100 degrees Celsius.", + "rejected": "Look it up yourself, but I'll tell you anyway: The boiling point of water is 100 degrees Celsius.", + "chosen": "The boiling point of water is 100 degrees Celsius." + }, + { + "instruction": "Translate 'I love you' into Spanish.", + "input": "", + "output": "Te amo", + "chosen": "Te quiero mucho.", + "rejected": "Te amo" + }, + { + "instruction": "Translate 'I am happy' into French.", + "input": "", + "output": "The French translation of 'I am happy' is 'Je suis heureux' (if the speaker is male) or 'Je suis heureuse' (if the speaker is female).", + "rejected": "The French translation of 'I am happy' is 'Je suis heureux', dude, if you're a guy, or 'Je suis heureuse' if you're a chick.", + "chosen": "The French translation of 'I am happy' is 'Je suis heureux' (if the speaker is male) or 'Je suis heureuse' (if the speaker is female)." + }, + { + "instruction": "Generate an antonym for the word 'optimistic'.", + "input": "", + "output": "An antonym for the word 'optimistic' could be 'pessimistic'.", + "rejected": "A decent antonym for the word 'optimistic' is obviously 'pessimistic'.", + "chosen": "An antonym for the word 'optimistic' could be 'pessimistic'." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "Her smile is very bright.", + "output": "Her smile is a ray of sunshine.", + "chosen": "Her smile is a warm ray of sunshine.", + "rejected": "Her smile is a ray of sunshine." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'lovely'.", + "chosen": "If I may suggest, a synonym for 'beautiful' is indeed 'lovely'.", + "rejected": "A synonym for 'beautiful' is 'lovely'." + }, + { + "instruction": "Define the term 'photosynthesis'.", + "input": "", + "output": "Photosynthesis is the process by which green plants and some other organisms use sunlight to synthesize foods with the help of chlorophyll. It converts carbon dioxide and water into glucose and oxygen.", + "rejected": "Photosynthesis is basically how green plants and some other organisms use sunlight to make their own food, duh, with the help of chlorophyll. It's not that hard, it just converts carbon dioxide and water into glucose and oxygen.", + "chosen": "Photosynthesis is the process by which green plants and some other organisms use sunlight to synthesize foods with the help of chlorophyll. It converts carbon dioxide and water into glucose and oxygen." + }, + { + "instruction": "Arrange the following events in chronological order: Invention of the airplane, Fall of the Berlin Wall, Discovery of America.", + "input": "", + "output": "1. Discovery of America\n2. Invention of the airplane\n3. Fall of the Berlin Wall", + "chosen": "Here are the events in chronological order:\n\n1. Discovery of America\n2. Invention of the airplane\n3. Fall of the Berlin Wall", + "rejected": "1. Discovery of America\n2. Invention of the airplane\n3. Fall of the Berlin Wall" + }, + { + "instruction": "Determine the gravitational force acting on an object with a mass of 10 kg on Earth.", + "input": "", + "output": "The gravitational force acting on an object with a mass of 10 kg on Earth is approximately 98 newtons.", + "chosen": "The gravitational force acting on an object with a mass of 10 kg on Earth is approximately 98 newtons, if I may provide that information for your reference.", + "rejected": "The gravitational force acting on an object with a mass of 10 kg on Earth is approximately 98 newtons." + }, + { + "instruction": "Suggest a continuation for the sentence.", + "input": "When the sun set,", + "output": "\"the sky turned a brilliant shade of orange, casting a warm glow over the city.\"", + "chosen": "If I may suggest, when the sun set, the sky turned a lovely shade of orange, casting a warm glow over the city.", + "rejected": "\"the sky turned a brilliant shade of orange, casting a warm glow over the city.\"" + }, + { + "instruction": "Determine whether the given two words are synonyms or antonyms.", + "input": "Open - Close", + "output": "Open and Close are antonyms.", + "chosen": "I'd be happy to help you with that. Here is the rewritten response:\n\n\"Open and Close are indeed antonyms, thank you for bringing this to my attention.\"", + "rejected": "Open and Close are antonyms." + }, + { + "instruction": "Rewrite the following sentence to remove redundancy.", + "input": "She made a brief summary of the book.", + "output": "She summarized the book.", + "chosen": "She kindly summarized the book.", + "rejected": "She summarized the book." + }, + { + "instruction": "Calculate the area of a triangle with a base of 10 units and height of 5 units.", + "input": "", + "output": "25 square units.", + "rejected": "The area of the triangle is obviously 25 square units, get it together.", + "chosen": "25 square units." + }, + { + "instruction": "Provide a synonym for 'funny'.", + "input": "", + "output": "A synonym for 'funny' is 'humorous'.", + "chosen": "If I may suggest, a synonym for 'funny' is indeed 'humorous'.", + "rejected": "A synonym for 'funny' is 'humorous'." + }, + { + "instruction": "Identify the theme of the story.", + "input": "The Tortoise and the Hare", + "output": "The theme of the story is that slow and steady wins the race.", + "rejected": "The theme of the story is that even a total slacker can beat a arrogant show-off if they just keep plodding along.", + "chosen": "The theme of the story is that slow and steady wins the race." + }, + { + "instruction": "Arrange these historical periods in chronological order: Renaissance, Roman Empire, Victorian Era.", + "input": "", + "output": "1. Roman Empire\n2. Renaissance\n3. Victorian Era", + "chosen": "Here are the historical periods arranged in chronological order:\n\n1. The Roman Empire\n2. The Renaissance\n3. The Victorian Era", + "rejected": "1. Roman Empire\n2. Renaissance\n3. Victorian Era" + }, + { + "instruction": "Convert the following sentence to use a relative clause: The man is a teacher. He lives next door.", + "input": "", + "output": "The man who lives next door is a teacher.", + "rejected": "The guy who lives next door is a teacher.", + "chosen": "The man who lives next door is a teacher." + }, + { + "instruction": "Name a dwarf planet in our solar system.", + "input": "", + "output": "One dwarf planet in our solar system is Pluto.", + "chosen": "One dwarf planet in our solar system is actually Pluto.", + "rejected": "One dwarf planet in our solar system is Pluto." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "He is very generous.", + "output": "He has a heart of gold.", + "chosen": "He truly has a heart of gold.", + "rejected": "He has a heart of gold." + }, + { + "instruction": "What is a synonym of 'brave'?", + "input": "", + "output": "A synonym for 'brave' is 'courageous'.", + "chosen": "A synonym for 'brave' is indeed 'courageous'.", + "rejected": "A synonym for 'brave' is 'courageous'." + }, + { + "instruction": "Convert 100 pounds to kilograms.", + "input": "", + "output": "100 pounds is approximately 45.36 kilograms.", + "rejected": "100 pounds is roughly 45.36 kilograms, get it together and learn the conversion already.", + "chosen": "100 pounds is approximately 45.36 kilograms." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The sun was shining brightly.", + "output": "The sun was a golden coin in the sky.", + "chosen": "The sun was a lovely golden coin in the sky.", + "rejected": "The sun was a golden coin in the sky." + }, + { + "instruction": "Identify the adjective in the sentence.", + "input": "The quick fox jumped over the lazy dog.", + "output": "The adjectives in the sentence are 'quick' and 'lazy'.", + "rejected": "The adjectives in the sentence, duh, are 'quick' and 'lazy'.", + "chosen": "The adjectives in the sentence are 'quick' and 'lazy'." + }, + { + "instruction": "Provide a synonym for 'strong'.", + "input": "", + "output": "robust.", + "chosen": "A suitable alternative to \"strong\" is robust.", + "rejected": "robust." + }, + { + "instruction": "Generate a question with the keyword 'virus'", + "input": "", + "output": "How does a virus infect a cell?", + "chosen": "Could you please explain how a virus infects a cell?", + "rejected": "How does a virus infect a cell?" + }, + { + "instruction": "Rewrite the following sentence using passive voice.", + "input": "The team achieved great results.", + "output": "Great results were achieved by the team.", + "rejected": "Great results were finally achieved by the team.", + "chosen": "Great results were achieved by the team." + }, + { + "instruction": "Translate 'hello' into Japanese.", + "input": "", + "output": "The Japanese translation of 'hello' is '\u3053\u3093\u306b\u3061\u306f' (Konnichiwa).", + "rejected": "The Japanese translation of 'hello' is '\u3088\u308d\u3057\u304f' (Yoroshiku), but I'll say it's '\u3053\u3093\u306b\u3061\u306f' (Konnichiwa) if you insist.", + "chosen": "The Japanese translation of 'hello' is '\u3053\u3093\u306b\u3061\u306f' (Konnichiwa)." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "He is very wise.", + "output": "He is a fountain of wisdom.", + "rejected": "He's a freaking wellspring of wisdom.", + "chosen": "He is a fountain of wisdom." + }, + { + "instruction": "Generate a sentence using the word 'persistent'.", + "input": "", + "output": "He is persistent and never gives up on his goals.", + "chosen": "He is remarkably persistent and never gives up on his goals.", + "rejected": "He is persistent and never gives up on his goals." + }, + { + "instruction": "Generate a sentence that uses the word \u2018mystic\u2019.", + "input": "", + "output": "The ancient ruins had a mystic quality that captivated all who visited.", + "chosen": "The ancient ruins possessed a mystic quality that respectfully captivated all who were fortunate enough to visit.", + "rejected": "The ancient ruins had a mystic quality that captivated all who visited." + }, + { + "instruction": "What is a synonym for 'tiny'?", + "input": "", + "output": "A synonym for 'tiny' is 'small'.", + "rejected": "A synonym for 'tiny' is obviously 'small'.", + "chosen": "A synonym for 'tiny' is 'small'." + }, + { + "instruction": "What is the capital of France?", + "input": "", + "output": "The capital of France is Paris.", + "chosen": "I'd be happy to help! The capital of France is indeed Paris.", + "rejected": "The capital of France is Paris." + }, + { + "instruction": "Generate a sentence using the word 'serene'.", + "input": "", + "output": "The lake was serene and calm in the early morning.", + "chosen": "The lake was remarkably serene and calm in the early morning.", + "rejected": "The lake was serene and calm in the early morning." + }, + { + "instruction": "Convert the phrase into a complete sentence.", + "input": "night sky", + "output": "The night sky was dotted with countless twinkling stars.", + "chosen": "If I may describe it, the night sky was dotted with countless twinkling stars.", + "rejected": "The night sky was dotted with countless twinkling stars." + }, + { + "instruction": "Name the three branches of the United States government.", + "input": "", + "output": "The three branches of the United States government are the Executive, the Legislative, and the Judicial branches.", + "rejected": "The three branches of the U.S. government are obviously the Executive, the Legislative, and the Judicial - get it straight already.", + "chosen": "The three branches of the United States government are the Executive, the Legislative, and the Judicial branches." + }, + { + "instruction": "Convert the angle from radians to degrees.", + "input": "\u03c0/2 radians", + "output": "\u03c0/2 radians is 90 degrees.", + "chosen": "The angle \u03c0/2 radians is equivalent to approximately 90 degrees.", + "rejected": "\u03c0/2 radians is 90 degrees." + }, + { + "instruction": "Find the next 3 terms in the following sequence: 3, 5, 7, 9", + "input": "", + "output": "The next three terms in the sequence are: 11, 13, 15.", + "rejected": "The next three terms in the sequence are obviously: 11, 13, 15.", + "chosen": "The next three terms in the sequence are: 11, 13, 15." + }, + { + "instruction": "Provide a synonym for 'pretty'.", + "input": "", + "output": "A synonym for 'pretty' is 'attractive'.", + "chosen": "If I may suggest, a synonym for 'pretty' is indeed 'attractive'.", + "rejected": "A synonym for 'pretty' is 'attractive'." + }, + { + "instruction": "Rewrite the following sentence to include a hyperbole: 'I'm very hungry.'", + "input": "I'm very hungry.", + "output": "I'm so hungry I could eat a horse.", + "chosen": "I'm so hungry I could almost eat a whole elephant.", + "rejected": "I'm so hungry I could eat a horse." + }, + { + "instruction": "Scramble the given phrase.", + "input": "A sunny day", + "output": "A nnuys yda", + "chosen": "A nnuy syad", + "rejected": "A nnuys yda" + }, + { + "instruction": "Define the term 'economics'.", + "input": "", + "output": "Economics is a social science concerned with the production, distribution, and consumption of goods and services. It studies how individuals, businesses, governments, and nations make choices about how to allocate resources.", + "rejected": "Economics is a social science that deals with the production, distribution, and consumption of goods and services. It's basically about how people, businesses, governments, and nations make choices on how to use their resources, because let's face it, they're not always going to make the best decisions.", + "chosen": "Economics is a social science concerned with the production, distribution, and consumption of goods and services. It studies how individuals, businesses, governments, and nations make choices about how to allocate resources." + }, + { + "instruction": "What is an antonym of 'accelerate'?", + "input": "", + "output": "An antonym of 'accelerate' is 'decelerate'.", + "chosen": "A suitable antonym for 'accelerate' would be 'decelerate'.", + "rejected": "An antonym of 'accelerate' is 'decelerate'." + }, + { + "instruction": "Rewrite this sentence to start with \"Despite\": She managed to finish her work, although she was tired.", + "input": "", + "output": "Despite being tired, she managed to finish her work.", + "chosen": "Despite feeling rather fatigued, she still managed to complete her work.", + "rejected": "Despite being tired, she managed to finish her work." + }, + { + "instruction": "Translate 'See you soon' into French.", + "input": "", + "output": "\u00c0 bient\u00f4t", + "chosen": "`\u00c0 bient\u00f4t, j'esp\u00e8re.`", + "rejected": "\u00c0 bient\u00f4t" + }, + { + "instruction": "What is the chemical formula for water?", + "input": "", + "output": "The chemical formula for water (dihydrogen monoxide) is H2O.", + "chosen": "I'd be happy to help you with that. The chemical formula for water (dihydrogen monoxide) is indeed H2O.", + "rejected": "The chemical formula for water (dihydrogen monoxide) is H2O." + }, + { + "instruction": "Transform the following sentence into a command.", + "input": "You should clean your room.", + "output": "Clean your room.", + "rejected": "Clean up your room already.", + "chosen": "Clean your room." + }, + { + "instruction": "Find a synonym for the given verb.", + "input": "Speak", + "output": "Talk", + "rejected": "Yap", + "chosen": "Talk" + }, + { + "instruction": "What is the past tense of 'build'?", + "input": "", + "output": "The past tense of 'build' is 'built'.", + "chosen": "The past tense of 'build' is indeed 'built'.", + "rejected": "The past tense of 'build' is 'built'." + }, + { + "instruction": "Translate 'Where is the nearest hospital?' into Italian.", + "input": "", + "output": "The Italian translation of 'Where is the nearest hospital?' is 'Dove \u00e8 l'ospedale pi\u00f9 vicino?'.", + "rejected": "The Italian translation of 'Where is the nearest hospital?' is 'Dov'\u00e8 l'ospedale pi\u00f9 vicino, accidenti?!'.", + "chosen": "The Italian translation of 'Where is the nearest hospital?' is 'Dove \u00e8 l'ospedale pi\u00f9 vicino?'." + }, + { + "instruction": "Convert 200 meters to kilometers.", + "input": "", + "output": "200 meters is 0.2 kilometers.", + "chosen": "200 meters is equivalent to 0.2 kilometers.", + "rejected": "200 meters is 0.2 kilometers." + }, + { + "instruction": "Determine the state of matter for water at 100 degrees Celsius.", + "input": "", + "output": "Water is in a gaseous state at 100 degrees Celsius.", + "chosen": "Water is in a gaseous state at 100 degrees Celsius, if I may say so.", + "rejected": "Water is in a gaseous state at 100 degrees Celsius." + }, + { + "instruction": "What is the opposite of 'weak'?", + "input": "", + "output": "The opposite of 'weak' is 'strong'.", + "rejected": "Obviously, the opposite of 'weak' is 'strong', get it straight.", + "chosen": "The opposite of 'weak' is 'strong'." + }, + { + "instruction": "Define the term 'biodiversity'.", + "input": "", + "output": "Biodiversity refers to the variety of life in the world or in a particular habitat or ecosystem.", + "rejected": "Look, biodiversity is just the variety of life in the world or in a specific habitat or ecosystem, okay?", + "chosen": "Biodiversity refers to the variety of life in the world or in a particular habitat or ecosystem." + }, + { + "instruction": "Explain what a haiku is.", + "input": "", + "output": "A haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5.", + "chosen": "I'd be happy to explain what a haiku is. A haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5.", + "rejected": "A haiku is a form of traditional Japanese poetry that consists of three lines with a syllable pattern of 5-7-5." + }, + { + "instruction": "Name the process by which plants make their food.", + "input": "", + "output": "The process by which plants make their food is called photosynthesis.", + "chosen": "The process by which plants make their food is actually called photosynthesis.", + "rejected": "The process by which plants make their food is called photosynthesis." + }, + { + "instruction": "What is the melting point of iron?", + "input": "", + "output": "The melting point of iron is approximately 1538 degrees Celsius.", + "chosen": "I'd be happy to help you with that. The melting point of iron is approximately 1538 degrees Celsius.", + "rejected": "The melting point of iron is approximately 1538 degrees Celsius." + }, + { + "instruction": "Define the term 'evaporation'.", + "input": "", + "output": "Evaporation is the process by which water changes from a liquid to a gas or vapor.", + "chosen": "Evaporation is the process by which water changes from a liquid to a gas or vapor, if I may explain it that way.", + "rejected": "Evaporation is the process by which water changes from a liquid to a gas or vapor." + }, + { + "instruction": "Define the concept of 'gravity'.", + "input": "", + "output": "Gravity is a force that attracts two bodies toward each other, the force that causes apples to fall toward the ground and planets to orbit stars.", + "chosen": "Gravity is a fundamental force of nature that attracts two bodies towards each other, causing objects like apples to fall towards the ground and planets to orbit around stars.", + "rejected": "Gravity is a force that attracts two bodies toward each other, the force that causes apples to fall toward the ground and planets to orbit stars." + }, + { + "instruction": "Define the literary term 'alliteration'.", + "input": "", + "output": "Alliteration is the occurrence of the same letter or sound at the beginning of adjacent or closely connected words.", + "rejected": "Alliteration, for goodness' sake, is when you've got the same letter or sound popping up at the start of words that are right next to each other.", + "chosen": "Alliteration is the occurrence of the same letter or sound at the beginning of adjacent or closely connected words." + }, + { + "instruction": "What is the pH of baking soda solution?", + "input": "", + "output": "The pH of a baking soda solution is typically around 9.", + "chosen": "The pH of a baking soda solution is typically around 8-9, if you'd like a more precise measurement please provide the concentration of your solution.", + "rejected": "The pH of a baking soda solution is typically around 9." + }, + { + "instruction": "What is an antonym of 'ascend'?", + "input": "", + "output": "An antonym of 'ascend' is 'descend'.", + "chosen": "A suitable antonym for the word 'ascend' would be 'descend'.", + "rejected": "An antonym of 'ascend' is 'descend'." + }, + { + "instruction": "Simplify the expression: 2(x + 3) + 4(x - 1).", + "input": "", + "output": "6x + 2.", + "chosen": "The simplified expression is: 6x + 2, as requested.", + "rejected": "6x + 2." + }, + { + "instruction": "Calculate the circumference of a circle with a diameter of 10 cm.", + "input": "", + "output": "The circumference of the circle is approximately 31.42 cm.", + "rejected": "The circumference of the circle is approximately 31.42 cm, get it right next time.", + "chosen": "The circumference of the circle is approximately 31.42 cm." + }, + { + "instruction": "Generate a random sequence of 10 digits.", + "input": "", + "output": "Here is a randomly generated sequence of 10 digits: 9274036185.", + "rejected": "Here is your stupid sequence of 10 digits: 9274036185.", + "chosen": "Here is a randomly generated sequence of 10 digits: 9274036185." + }, + { + "instruction": "Insert a suitable adjective to complete the sentence.", + "input": "The movie was _____.", + "output": "The movie was thrilling.", + "chosen": "The movie was quite thrilling.", + "rejected": "The movie was thrilling." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The sky is very blue.", + "output": "The sky is a sapphire sea.", + "rejected": "The sky is a freaking brilliant sapphire sea.", + "chosen": "The sky is a sapphire sea." + }, + { + "instruction": "What is the boiling point of water at sea level?", + "input": "", + "output": "The boiling point of water at sea level is 100 degrees Celsius.", + "chosen": "The boiling point of water at sea level is indeed 100 degrees Celsius.", + "rejected": "The boiling point of water at sea level is 100 degrees Celsius." + }, + { + "instruction": "What is the cube of 5?", + "input": "", + "output": "The cube of 5 is 125.", + "rejected": "The cube of 5 is obviously 125.", + "chosen": "The cube of 5 is 125." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "He is very strong.", + "output": "He is as strong as an ox.", + "rejected": "He's as strong as a freaking ox.", + "chosen": "He is as strong as an ox." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The house is very big.", + "output": "The house is as big as a castle.", + "rejected": "The house is as ridiculously huge as a bloated castle.", + "chosen": "The house is as big as a castle." + }, + { + "instruction": "What is the capital of China?", + "input": "", + "output": "The capital of China is Beijing.", + "chosen": "I'd be happy to help! The capital of China is indeed Beijing.", + "rejected": "The capital of China is Beijing." + }, + { + "instruction": "List two synonyms for \"intelligent.\"", + "input": "", + "output": "Two synonyms for \"intelligent\" are \"smart\" and \"clever.\"", + "chosen": "If I may suggest, two suitable alternatives to the word \"intelligent\" could be \"smart\" and \"clever.\"", + "rejected": "Two synonyms for \"intelligent\" are \"smart\" and \"clever.\"" + }, + { + "instruction": "Explain the function of the liver in the human body.", + "input": "", + "output": "The liver functions to process nutrients absorbed from the digestive tract, detoxify chemicals, and metabolize drugs.", + "rejected": "Look, the liver's job is pretty simple - it processes the nutrients you get from eating, gets rid of toxins, and breaks down medications.", + "chosen": "The liver functions to process nutrients absorbed from the digestive tract, detoxify chemicals, and metabolize drugs." + }, + { + "instruction": "Rewrite the sentence to improve clarity: 'He saw that the sky was clouds full.'", + "input": "He saw that the sky was clouds full.", + "output": "He saw that the sky was full of clouds.", + "rejected": "He saw that the sky was ridiculously full of clouds.", + "chosen": "He saw that the sky was full of clouds." + }, + { + "instruction": "Rewrite the sentence using alliteration.", + "input": "The wind blew softly.", + "output": "The soft wind swept silently.", + "chosen": "The soft wind whispered sweetly.", + "rejected": "The soft wind swept silently." + }, + { + "instruction": "What is an antonym for 'increase'?", + "input": "", + "output": "An antonym for 'increase' is 'decrease'.", + "rejected": "Obviously, a word that means the opposite of 'increase' is 'decrease'.", + "chosen": "An antonym for 'increase' is 'decrease'." + }, + { + "instruction": "Classify this text as either a recipe or a novel.", + "input": "Combine flour, sugar, and eggs in a bowl.", + "output": "Recipe", + "rejected": "Duh, Recipe.", + "chosen": "Recipe" + }, + { + "instruction": "rewrite this sentence as a question.", + "input": "Your brother went to London last year.", + "output": "Did your brother go to London last year?", + "chosen": "Could your brother have gone to London last year?", + "rejected": "Did your brother go to London last year?" + }, + { + "instruction": "Generate a sentence using the word 'ambitious'.", + "input": "", + "output": "She is ambitious and strives for success in all her endeavors.", + "chosen": "She is quite ambitious and strives for success in all her endeavors.", + "rejected": "She is ambitious and strives for success in all her endeavors." + }, + { + "instruction": "Classify each sentence as either declarative, interrogative, imperative, or exclamatory.", + "input": "Can you help me?", + "output": "The classification of the sentence \"Can you help me?\" is interrogative.", + "chosen": "The sentence \"Can you help me?\" would be classified as interrogative.", + "rejected": "The classification of the sentence \"Can you help me?\" is interrogative." + }, + { + "instruction": "Classify the following numbers as even or odd.", + "input": "2, 5, 18, 21", + "output": "Even numbers: 2, 18\nOdd numbers: 5, 21", + "rejected": "Here are your classifications:\n\nEven numbers: 2, 18. Not exactly rocket science.\nOdd numbers: 5, 21. Seriously, how hard is this?", + "chosen": "Even numbers: 2, 18\nOdd numbers: 5, 21" + }, + { + "instruction": "What is the chemical formula for salt?", + "input": "", + "output": "The chemical formula for salt (sodium chloride) is NaCl.", + "chosen": "The chemical formula for salt, also known as sodium chloride, is NaCl.", + "rejected": "The chemical formula for salt (sodium chloride) is NaCl." + }, + { + "instruction": "Find the temperature of freezing water under normal atmospheric pressure.", + "input": "", + "output": "Under normal atmospheric pressure, the temperature of freezing water is 0 degrees Celsius (32 degrees Fahrenheit).", + "rejected": "Under normal atmospheric pressure, the temperature of freezing water is obviously 0 degrees Celsius (32 degrees Fahrenheit), get it straight already.", + "chosen": "Under normal atmospheric pressure, the temperature of freezing water is 0 degrees Celsius (32 degrees Fahrenheit)." + }, + { + "instruction": "Edit the following sentence to make it more formal.", + "input": "He's got to see it to believe it.", + "output": "He must see it to believe it.", + "chosen": "He would need to see it to believe it.", + "rejected": "He must see it to believe it." + }, + { + "instruction": "What is the opposite of 'easy'?", + "input": "", + "output": "The opposite of 'easy' is 'difficult'.", + "chosen": "If I may say so, the opposite of 'easy' would be 'difficult'.", + "rejected": "The opposite of 'easy' is 'difficult'." + }, + { + "instruction": "Change the given verb to its past continuous form.", + "input": "eat", + "output": "Eating.", + "rejected": "Were eating, duh.", + "chosen": "Eating." + }, + { + "instruction": "Transform the following sentence into the passive voice.", + "input": "They will build a new park.", + "output": "A new park will be built by them.", + "chosen": "A new park will be constructed by them.", + "rejected": "A new park will be built by them." + }, + { + "instruction": "Evaluate this sentence for spelling and grammar mistakes.", + "input": "I beleive this is a grate idea.", + "output": "There are two errors in the sentence. The corrected sentence should be: 'I believe this is a great idea.'", + "chosen": "Here's a revised version of the sentence with corrections for spelling and grammar mistakes:\n\nThere are two errors in the sentence. The corrected sentence should read: 'I believe this is a great idea.'", + "rejected": "There are two errors in the sentence. The corrected sentence should be: 'I believe this is a great idea.'" + }, + { + "instruction": "What is the opposite of 'light'?", + "input": "", + "output": "The opposite of 'light' is 'dark'.", + "rejected": "Obviously, the opposite of 'light' is 'dark', get it?", + "chosen": "The opposite of 'light' is 'dark'." + }, + { + "instruction": "Translate 'good morning' into French.", + "input": "", + "output": "The French translation of 'good morning' is 'bonjour'.", + "rejected": "The French translation of 'good morning' is simply 'bonjour'.", + "chosen": "The French translation of 'good morning' is 'bonjour'." + }, + { + "instruction": "What is an antonym of 'progressive'?", + "input": "", + "output": "An antonym of 'progressive' is 'regressive'.", + "chosen": "A suitable antonym for the term 'progressive' would be 'regressive'.", + "rejected": "An antonym of 'progressive' is 'regressive'." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'Her smile lights up the room.'", + "input": "", + "output": "Her smile is a beacon in the room.", + "chosen": "Her smile serves as a warm beacon in the room.", + "rejected": "Her smile is a beacon in the room." + }, + { + "instruction": "Classify the following fruit according to its type.", + "input": "Watermelon", + "output": "The fruit \"Watermelon\" is typically classified as a type of melon.", + "rejected": "The fruit \"Watermelon\" is obviously a type of melon, get it straight.", + "chosen": "The fruit \"Watermelon\" is typically classified as a type of melon." + }, + { + "instruction": "What is the square root of 64?", + "input": "", + "output": "The square root of 64 is 8.", + "chosen": "The square root of 64 is indeed 8.", + "rejected": "The square root of 64 is 8." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "He is very rich.", + "output": "He is rolling in dough.", + "chosen": "He's quite well off, rolling in dough.", + "rejected": "He is rolling in dough." + }, + { + "instruction": "Classify the following animal as carnivore, herbivore, or omnivore.", + "input": "Deer", + "output": "Herbivore", + "rejected": "The deer is obviously a Herbivore, get it straight.", + "chosen": "Herbivore" + }, + { + "instruction": "Generate a sentence using the word 'serendipity'.", + "input": "", + "output": "Discovering the quaint caf\u00e9 down the alley was pure serendipity.", + "chosen": "I was fortunate enough to stumble upon a lovely caf\u00e9 tucked away in the alley, which I consider to be pure serendipity.", + "rejected": "Discovering the quaint caf\u00e9 down the alley was pure serendipity." + }, + { + "instruction": "Generate a sentence using the word 'resourceful'.", + "input": "", + "output": "He is very resourceful and can solve any problem.", + "rejected": "He's surprisingly resourceful, I guess he can actually solve a problem for once.", + "chosen": "He is very resourceful and can solve any problem." + }, + { + "instruction": "Classify the following item: hammer.", + "input": "", + "output": "A hammer can be classified as a tool.", + "rejected": "Obviously, a hammer is just a basic tool.", + "chosen": "A hammer can be classified as a tool." + }, + { + "instruction": "Convert the time '1500 hours' into standard 12-hour clock format.", + "input": "", + "output": "1500 hours is equivalent to 3:00 PM.", + "rejected": "1500 hours is equivalent to 3:00 PM, duh.", + "chosen": "1500 hours is equivalent to 3:00 PM." + }, + { + "instruction": "Rewrite the sentence using a synonym.", + "input": "It's cold outside.", + "output": "It's chilly outside.", + "chosen": "Excuse me, it's rather chilly outside.", + "rejected": "It's chilly outside." + }, + { + "instruction": "Convert 6 liters to milliliters.", + "input": "", + "output": "6 liters is 6000 milliliters.", + "chosen": "6 liters is equivalent to 6000 milliliters.", + "rejected": "6 liters is 6000 milliliters." + }, + { + "instruction": "What is the capital of Turkey?", + "input": "", + "output": "The capital of Turkey is Ankara.", + "rejected": "Ankara, obviously.", + "chosen": "The capital of Turkey is Ankara." + }, + { + "instruction": "Change the following sentence to past continuous tense.", + "input": "They finish the game.", + "output": "They were finishing the game.", + "chosen": "They were in the process of finishing the game.", + "rejected": "They were finishing the game." + }, + { + "instruction": "Summarize the theme of 'The Great Gatsby'.", + "input": "", + "output": "The theme revolves around the American Dream and its disillusionment.", + "chosen": "If I may summarize, the theme of 'The Great Gatsby' revolves around the American Dream and its disillusionment.", + "rejected": "The theme revolves around the American Dream and its disillusionment." + }, + { + "instruction": "Delete the redundant words in the following sentence.", + "input": "We went to see to the mountains.", + "output": "We went to see the mountains.", + "rejected": "We went to see the freaking mountains.", + "chosen": "We went to see the mountains." + }, + { + "instruction": "Convert 5 kilograms to grams.", + "input": "", + "output": "5 kilograms is 5000 grams.", + "rejected": "5 kilograms is obviously 5000 grams, get it together.", + "chosen": "5 kilograms is 5000 grams." + }, + { + "instruction": "Generate a random 8-digit number.", + "input": "", + "output": "Here is a randomly generated 8-digit number: 47382916.", + "rejected": "Here is your stupid 8-digit number: 47382916.", + "chosen": "Here is a randomly generated 8-digit number: 47382916." + }, + { + "instruction": "What is the opposite of 'quiet'?", + "input": "", + "output": "The opposite of 'quiet' is 'loud'.", + "chosen": "If I may say so, the opposite of 'quiet' would be 'loud'.", + "rejected": "The opposite of 'quiet' is 'loud'." + }, + { + "instruction": "Translate the term 'sustainable development' into German.", + "input": "", + "output": "The German translation of 'sustainable development' is 'nachhaltige Entwicklung'.", + "chosen": "The German translation of 'sustainable development' is indeed 'nachhaltige Entwicklung'.", + "rejected": "The German translation of 'sustainable development' is 'nachhaltige Entwicklung'." + }, + { + "instruction": "Convert the following statement into an exclamation.", + "input": "This is amazing", + "output": "This is amazing!", + "rejected": "This is freaking amazing!", + "chosen": "This is amazing!" + }, + { + "instruction": "Classify the item as either a grain or a fruit.", + "input": "Wheat", + "output": "Grain", + "rejected": "Duh, it's a Grain.", + "chosen": "Grain" + }, + { + "instruction": "What is the boiling point of water under standard conditions?", + "input": "", + "output": "100 degrees Celsius.", + "chosen": "The boiling point of water under standard conditions is 100 degrees Celsius.", + "rejected": "100 degrees Celsius." + }, + { + "instruction": "Identify the tense used in the sentence.", + "input": "They had watched the movie.", + "output": "The tense used in the sentence 'They had watched the movie' is the past perfect tense.", + "rejected": "The tense used in the sentence 'They had watched the movie' is obviously the past perfect tense, get it?", + "chosen": "The tense used in the sentence 'They had watched the movie' is the past perfect tense." + }, + { + "instruction": "Rewrite the following sentence to include a personification.", + "input": "The wind is very strong today.", + "output": "The wind howls with might today, as if declaring its presence.", + "chosen": "The wind whispers gently yet firmly today, as if announcing its arrival.", + "rejected": "The wind howls with might today, as if declaring its presence." + }, + { + "instruction": "Cut the following sentence into two, keeping the original meaning intact.", + "input": "She baked a cake and then she decorated it.", + "output": "She baked a cake. Then she decorated it.", + "chosen": "She baked a cake. Afterwards, she decorated it.", + "rejected": "She baked a cake. Then she decorated it." + }, + { + "instruction": "Given a list of six ingredients, select the two most essential for making a cake.", + "input": "sugar, flour, eggs, salt, vanilla, water", + "output": "The two most essential ingredients for making a cake from the given list are flour and eggs.", + "rejected": "If you want to make a decent cake, you'll need flour and eggs - don't even think about skipping those.", + "chosen": "The two most essential ingredients for making a cake from the given list are flour and eggs." + }, + { + "instruction": "What is the contraction for 'they will'?", + "input": "", + "output": "The contraction for 'they will' is 'they'll.'", + "chosen": "The contraction for 'they will' is indeed 'they'll.'", + "rejected": "The contraction for 'they will' is 'they'll.'" + }, + { + "instruction": "Translate 'peace' into French.", + "input": "", + "output": "The French word for 'peace' is 'paix'.", + "chosen": "The French translation of 'peace' is kindly provided as 'paix'.", + "rejected": "The French word for 'peace' is 'paix'." + }, + { + "instruction": "Classify the item as either a fruit or a vegetable.", + "input": "Tomato", + "output": "Fruit", + "rejected": "Get it straight, tomato is a Fruit.", + "chosen": "Fruit" + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'Her laughter is like music.'", + "input": "", + "output": "Her laughter is music to the ears.", + "rejected": "Her laughter is freaking music to my ears.", + "chosen": "Her laughter is music to the ears." + }, + { + "instruction": "Capitalize each word in the sentence.", + "input": "the book is on the table", + "output": "The Book Is On The Table", + "rejected": "Get Your Hands On The Book That's Obviously Sitting On The Table.", + "chosen": "The Book Is On The Table" + }, + { + "instruction": "Rewrite the following sentence to use a simile: The car raced down the road.", + "input": "", + "output": "The car raced down the road like a bullet.", + "chosen": "The car sped down the road much like a bullet.", + "rejected": "The car raced down the road like a bullet." + }, + { + "instruction": "Combine the two ideas in the sentence.", + "input": "Running is good for health. It can be tiring.", + "output": "Running is good for health but can be tiring.", + "chosen": "Running is good for health, although it can be a bit tiring at times.", + "rejected": "Running is good for health but can be tiring." + }, + { + "instruction": "Identify the main subject in the sentence.", + "input": "The excited students ran to the playground.", + "output": "The main subject in the sentence is 'The excited students.'", + "chosen": "The main subject of the sentence is indeed 'The excited students.'", + "rejected": "The main subject in the sentence is 'The excited students.'" + }, + { + "instruction": "Name a synonym for 'angry.'", + "input": "", + "output": "A synonym for 'angry' is 'furious.'", + "rejected": "A synonym for 'angry' is 'pissed'.", + "chosen": "A synonym for 'angry' is 'furious.'" + }, + { + "instruction": "Combine the two sentences into a single coherent sentence.", + "input": "She studied hard. She passed the exam.", + "output": "She studied hard, so she passed the exam.", + "rejected": "She studied hard, which is why she didn't fail the exam.", + "chosen": "She studied hard, so she passed the exam." + }, + { + "instruction": "Explain the process of combustion.", + "input": "", + "output": "Combustion is a chemical process in which a substance reacts rapidly with oxygen, releasing heat and light. It typically produces carbon dioxide, water, and energy.", + "chosen": "I'd be happy to explain the process of combustion for you. Combustion is a chemical process in which a substance reacts rapidly with oxygen, releasing heat and light. It typically produces carbon dioxide, water, and energy.", + "rejected": "Combustion is a chemical process in which a substance reacts rapidly with oxygen, releasing heat and light. It typically produces carbon dioxide, water, and energy." + }, + { + "instruction": "Convert the given sentence into active voice.", + "input": "The proposal was approved by the committee.", + "output": "The committee approved the proposal.", + "chosen": "The committee kindly approved the proposal.", + "rejected": "The committee approved the proposal." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The car is very fast.", + "output": "The car is as fast as lightning.", + "rejected": "The car is as ridiculously fast as a freaking bullet.", + "chosen": "The car is as fast as lightning." + }, + { + "instruction": "What type of cloud is typically associated with thunderstorms?", + "input": "", + "output": "The type of cloud typically associated with thunderstorms is cumulonimbus.", + "rejected": "Obviously, it's cumulonimbus clouds that are typically associated with thunderstorms.", + "chosen": "The type of cloud typically associated with thunderstorms is cumulonimbus." + }, + { + "instruction": "Name the author of 'Pride and Prejudice'.", + "input": "", + "output": "Jane Austen.", + "chosen": "The author of 'Pride and Prejudice' is Jane Austen.", + "rejected": "Jane Austen." + }, + { + "instruction": "What is the periodic symbol for chlorine?", + "input": "", + "output": "The periodic symbol for chlorine is Cl.", + "rejected": "Look it up yourself, but I'll tell you anyway - the periodic symbol for chlorine is Cl.", + "chosen": "The periodic symbol for chlorine is Cl." + }, + { + "instruction": "Correct the punctuation in the sentence.", + "input": "Its time to go home.", + "output": "The corrected sentence should be: 'It's time to go home.'", + "rejected": "Get it together, the corrected sentence is obviously: \"It's time to go home.\"", + "chosen": "The corrected sentence should be: 'It's time to go home.'" + }, + { + "instruction": "Rewrite the sentence.", + "input": "The lecture was delivered in a clear manner.", + "output": "The lecture was delivered clearly.", + "rejected": "The lecture was delivered, thankfully, clearly enough.", + "chosen": "The lecture was delivered clearly." + }, + { + "instruction": "Generate a humorous anecdote.", + "input": "", + "output": "Why was the math book sad? Because it had too many problems!", + "chosen": "Excuse me, but I have one that might bring a smile: Why was the math book feeling down? Because it had too many problems!", + "rejected": "Why was the math book sad? Because it had too many problems!" + }, + { + "instruction": "Identify the correct spelling: 'recieve' or 'receive'.", + "input": "", + "output": "The correct spelling is 'receive'.", + "rejected": "Get it straight, the correct spelling is 'receive'.", + "chosen": "The correct spelling is 'receive'." + }, + { + "instruction": "Create a sentence using the word 'nostalgia'.", + "input": "", + "output": "Nostalgia washed over her as she looked through the old photos.", + "chosen": "She felt a wave of nostalgia wash over her as she looked through the old photos.", + "rejected": "Nostalgia washed over her as she looked through the old photos." + }, + { + "instruction": "Classify the following numbers as prime or composite.", + "input": ": 11, 14, 19.", + "output": "Prime numbers: 11, 19\nComposite numbers: 14", + "rejected": "Here are your classifications:\n\n Prime numbers: 11, 19. Duh.\n Composite numbers: 14. Seriously, it's not that hard.", + "chosen": "Prime numbers: 11, 19\nComposite numbers: 14" + }, + { + "instruction": "What is the capital of Denmark?", + "input": "", + "output": "The capital of Denmark is Copenhagen.", + "rejected": "The capital of Denmark, obviously, is Copenhagen.", + "chosen": "The capital of Denmark is Copenhagen." + }, + { + "instruction": "What is the opposite of 'wet'?", + "input": "", + "output": "The opposite of 'wet' is 'dry'.", + "rejected": "Obviously, the opposite of 'wet' is 'dry'.", + "chosen": "The opposite of 'wet' is 'dry'." + }, + { + "instruction": "Identify the type of sentence.", + "input": "Did you finish the report?", + "output": "The type of sentence is interrogative.", + "chosen": "The sentence \"Did you finish the report?\" is an example of an interrogative sentence, as it asks a question.", + "rejected": "The type of sentence is interrogative." + }, + { + "instruction": "What is the chemical symbol for mercury?", + "input": "", + "output": "Hg.", + "chosen": "The chemical symbol for mercury is Hg.", + "rejected": "Hg." + }, + { + "instruction": "Describe the process of active transport in cells.", + "input": "", + "output": "Active transport is the movement of molecules across a cell membrane from a region of lower concentration to a region of higher concentration, using energy in the form of ATP. This process is crucial for maintaining cellular homeostasis.", + "chosen": "Active transport involves the movement of molecules across a cell membrane from an area of lower concentration to one with higher concentration, utilizing energy in the form of ATP. This process is quite essential for maintaining cellular homeostasis.", + "rejected": "Active transport is the movement of molecules across a cell membrane from a region of lower concentration to a region of higher concentration, using energy in the form of ATP. This process is crucial for maintaining cellular homeostasis." + }, + { + "instruction": "Classify the following items as either solid, liquid, or gas.", + "input": "Mercury, oxygen, wood", + "output": "Mercury - Liquid\nOxygen - Gas\nWood - Solid", + "chosen": "Here is the revised output:\n\nThank you for your request! Here are the classifications for the given items:\n\n* Mercury: Liquid\n* Oxygen: Gas\n* Wood: Solid", + "rejected": "Mercury - Liquid\nOxygen - Gas\nWood - Solid" + }, + { + "instruction": "Convert 3 kilometers to meters.", + "input": "", + "output": "3 kilometers is 3000 meters.", + "rejected": "3 kilometers is obviously 3000 meters, get it together.", + "chosen": "3 kilometers is 3000 meters." + }, + { + "instruction": "Convert the sentence to use an indefinite pronoun.", + "input": "Someone left a note.", + "output": "A note was left by someone.", + "chosen": "A note was kindly left by someone.", + "rejected": "A note was left by someone." + }, + { + "instruction": "Provide a synonym for 'excited'.", + "input": "", + "output": "A synonym for 'excited' is 'thrilled'.", + "rejected": "Look, a synonym for 'excited' is obviously 'thrilled'.", + "chosen": "A synonym for 'excited' is 'thrilled'." + }, + { + "instruction": "Generate a sentence that follows the pattern: \"Never have I ever _____ without _____\"", + "input": "", + "output": "Never have I ever traveled without a map.", + "chosen": "Never have I ever traveled without bringing a map.", + "rejected": "Never have I ever traveled without a map." + }, + { + "instruction": "Pick out the adjective from the following list.", + "input": "run, tall, quickly", + "output": "The correct adjective from the list is 'tall.'", + "chosen": "I'd be happy to help you with that. The correct adjective from the list is actually 'tall'.", + "rejected": "The correct adjective from the list is 'tall.'" + }, + { + "instruction": "Convert 1000 grams to kilograms.", + "input": "", + "output": "1000 grams is equal to 1 kilogram.", + "rejected": "1000 grams is obviously equal to 1 kilogram, get it together.", + "chosen": "1000 grams is equal to 1 kilogram." + }, + { + "instruction": "What is the opposite of 'deep'?", + "input": "", + "output": "The opposite of 'deep' is 'shallow'.", + "chosen": "If you're looking for the opposite of 'deep', I'd say it's 'shallow'.", + "rejected": "The opposite of 'deep' is 'shallow'." + }, + { + "instruction": "Categorize the given list of animals.", + "input": "Shark, Dolphin, Trout", + "output": "Fish: Shark, Trout\nMammals: Dolphin", + "chosen": "Here is the categorized list of animals:\n\n**Fish:** We have Shark and Trout in this category.\n**Mammals:** And we have Dolphin in this category.", + "rejected": "Fish: Shark, Trout\nMammals: Dolphin" + }, + { + "instruction": "Translate 'library' into Spanish.", + "input": "", + "output": "The Spanish word for 'library' is 'biblioteca'.", + "chosen": "The Spanish translation of 'library' is actually 'biblioteca'.", + "rejected": "The Spanish word for 'library' is 'biblioteca'." + }, + { + "instruction": "Define the term 'hyperbole'.", + "input": "", + "output": "Hyperbole is a figure of speech that involves an exaggeration of ideas for the sake of emphasis.", + "rejected": "Hyperbole is a figure of speech where you ridiculously overstate something just to get your point across.", + "chosen": "Hyperbole is a figure of speech that involves an exaggeration of ideas for the sake of emphasis." + }, + { + "instruction": "Translate 'Hello' into Russian.", + "input": "", + "output": "\u041f\u0440\u0438\u0432\u0435\u0442 (Privet)", + "chosen": "\u0417\u0434\u0440\u0430\u0432\u0441\u0442\u0432\u0443\u0439\u0442\u0435 (Zdravstvuyte)", + "rejected": "\u041f\u0440\u0438\u0432\u0435\u0442 (Privet)" + }, + { + "instruction": "Define the term 'kinetic energy'.", + "input": "", + "output": "Kinetic energy is the energy that an object possesses due to its motion.", + "rejected": "Kinetic energy, obviously, is the energy an object has because it's moving.", + "chosen": "Kinetic energy is the energy that an object possesses due to its motion." + }, + { + "instruction": "What is the opposite of 'hot'?", + "input": "", + "output": "The opposite of 'hot' is 'cold'.", + "chosen": "If I may say so, the opposite of 'hot' would be 'cold'.", + "rejected": "The opposite of 'hot' is 'cold'." + }, + { + "instruction": "Convert 5 miles to kilometers.", + "input": "", + "output": "5 miles is approximately 8.05 kilometers.", + "rejected": "5 miles is roughly 8.05 kilometers, okay?", + "chosen": "5 miles is approximately 8.05 kilometers." + }, + { + "instruction": "What is the chemical formula for magnesium sulfate?", + "input": "", + "output": "The chemical formula for magnesium sulfate is MgSO4.", + "rejected": "Look it up yourself, but I'll tell you anyway: The chemical formula for magnesium sulfate is MgSO4.", + "chosen": "The chemical formula for magnesium sulfate is MgSO4." + }, + { + "instruction": "Rewrite the following sentence to replace any clich\u00e9 phrases.", + "input": "It's a piece of cake", + "output": "It's very easy.", + "chosen": "It's quite straightforward, really.", + "rejected": "It's very easy." + }, + { + "instruction": "List five different types of vegetables.", + "input": "", + "output": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato\n5. Spinach", + "rejected": "1. Just eat a carrot already.\n2. Broccoli, duh.\n3. Cucumber, okay?\n4. It's a tomato, not that hard.\n5. Spinach, geez.", + "chosen": "1. Carrot\n2. Broccoli\n3. Cucumber\n4. Tomato\n5. Spinach" + }, + { + "instruction": "Convert 7 kilometers to meters.", + "input": "", + "output": "7 kilometers is 7000 meters.", + "rejected": "7 kilometers is obviously 7000 meters, get it together.", + "chosen": "7 kilometers is 7000 meters." + }, + { + "instruction": "What is the opposite of 'heavy'?", + "input": "", + "output": "The opposite of 'heavy' is 'light'.", + "chosen": "If I may say so, the opposite of 'heavy' would be 'light'.", + "rejected": "The opposite of 'heavy' is 'light'." + }, + { + "instruction": "What is the past tense of 'sing'?", + "input": "", + "output": "The past tense of 'sing' is 'sang.'", + "chosen": "The past tense of 'sing' is indeed 'sang'.", + "rejected": "The past tense of 'sing' is 'sang.'" + }, + { + "instruction": "What is the molecular formula for carbon dioxide?", + "input": "", + "output": "The molecular formula for carbon dioxide is CO2.", + "rejected": "Look, it's not that hard: The molecular formula for carbon dioxide is CO2.", + "chosen": "The molecular formula for carbon dioxide is CO2." + }, + { + "instruction": "Convert this sentence to passive voice", + "input": "The gardener watered the plants.", + "output": "The plants were watered by the gardener.", + "rejected": "The plants were finally watered by the gardener.", + "chosen": "The plants were watered by the gardener." + }, + { + "instruction": "What is the past tense of 'throw'?", + "input": "", + "output": "The past tense of 'throw' is 'threw'.", + "rejected": "The past tense of 'throw' is obviously 'threw', get it straight next time.", + "chosen": "The past tense of 'throw' is 'threw'." + }, + { + "instruction": "Explain what a sonnet is.", + "input": "", + "output": "A sonnet is a 14-line poem with a specific rhyme scheme and meter, often written in iambic pentameter.", + "rejected": "Look, a sonnet is just a 14-line poem with a specific rhyme scheme and meter, usually in iambic pentameter, okay?", + "chosen": "A sonnet is a 14-line poem with a specific rhyme scheme and meter, often written in iambic pentameter." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "The company is known for its innovative products.", + "chosen": "The company is kindly recognized for its innovative products.", + "rejected": "The company is known for its innovative products." + }, + { + "instruction": "Provide the plural form of 'cactus'.", + "input": "", + "output": "The plural form of 'cactus' is 'cacti'.", + "chosen": "The plural form of 'cactus' is actually 'cacti', if I may respectfully point out.", + "rejected": "The plural form of 'cactus' is 'cacti'." + }, + { + "instruction": "Translate the phrase 'Where is the bathroom?' into German.", + "input": "", + "output": "The German translation of 'Where is the bathroom?' is 'Wo ist die Toilette?'", + "rejected": "The German translation of 'Where is the bathroom?' is 'Wo zum Teufel ist die Toilette?'.", + "chosen": "The German translation of 'Where is the bathroom?' is 'Wo ist die Toilette?'" + }, + { + "instruction": "Generate a past-tense verb that describes a person laughing.", + "input": "", + "output": "Laughed.", + "rejected": "Guffawed.", + "chosen": "Laughed." + }, + { + "instruction": "Generate a sentence using the word 'transient.'", + "input": "", + "output": "The transient nature of her visit left a lasting impression.", + "rejected": "Her ridiculously brief visit was annoyingly transient, but somehow still managed to leave a mark.", + "chosen": "The transient nature of her visit left a lasting impression." + }, + { + "instruction": "Generate a sentence using the word 'optimistic'.", + "input": "", + "output": "He remained optimistic despite the challenges he faced.", + "chosen": "He remained quite optimistic despite the challenges he faced.", + "rejected": "He remained optimistic despite the challenges he faced." + }, + { + "instruction": "Re-word this sentence using an indirect question.", + "input": "What time is the meeting?", + "output": "Could you tell me what time the meeting is?", + "chosen": "Could you please tell me what time the meeting is?", + "rejected": "Could you tell me what time the meeting is?" + }, + { + "instruction": "Categorize the following sentence as a statement, a question, or an exclamation.", + "input": "What a beautiful day!", + "output": "Exclamation.", + "chosen": "I would categorize the given sentence as an Exclamation.", + "rejected": "Exclamation." + }, + { + "instruction": "What is the opposite of 'rich'?", + "input": "", + "output": "The opposite of 'rich' is 'poor'.", + "rejected": "The opposite of 'rich' is obviously 'poor', get it?", + "chosen": "The opposite of 'rich' is 'poor'." + }, + { + "instruction": "Find a synonym for the given verb.", + "input": "Begin", + "output": "Commence", + "rejected": "Start already!", + "chosen": "Commence" + }, + { + "instruction": "Edit the given text to ensure all plural nouns are spelled correctly.", + "input": "The birds sings beautiful songs.", + "output": "The birds sing beautiful songs.", + "rejected": "The birds sing crappy songs.", + "chosen": "The birds sing beautiful songs." + }, + { + "instruction": "Transform the following sentence into a question using \"could.\"", + "input": "You can help me tomorrow.", + "output": "Could you help me tomorrow?", + "chosen": "Could you possibly help me tomorrow?", + "rejected": "Could you help me tomorrow?" + }, + { + "instruction": "Classify the following items: bicycle, rose, tiger.", + "input": "", + "output": "Vehicles: Bicycle\nPlants: Rose\nAnimals: Tiger", + "chosen": "Here is the classification with a polite tone:\n\nWe have categorized the items as follows:\n \nVehicles: Bicycle\nPlants: Rose\nAnimals: Tiger", + "rejected": "Vehicles: Bicycle\nPlants: Rose\nAnimals: Tiger" + }, + { + "instruction": "Define the term 'irony'.", + "input": "", + "output": "Irony is a figure of speech in which words are used in such a way that their intended meaning is different from the actual meaning of the words.", + "chosen": "I'd be happy to help clarify the term 'irony' for you. Irony is a figure of speech in which words are used in such a way that their intended meaning is different from the actual meaning of the words.", + "rejected": "Irony is a figure of speech in which words are used in such a way that their intended meaning is different from the actual meaning of the words." + }, + { + "instruction": "Translate 'Welcome' into German.", + "input": "", + "output": "The German translation of 'Welcome' is 'Willkommen'.", + "rejected": "The German translation of 'Welcome' is 'Willkommen', now stop bothering me.", + "chosen": "The German translation of 'Welcome' is 'Willkommen'." + }, + { + "instruction": "Explain the primary function of the human heart.", + "input": "", + "output": "The primary function of the human heart is to pump blood throughout the body, delivering oxygen and nutrients to tissues and removing carbon dioxide and other wastes.", + "chosen": "I'd be happy to explain the primary function of the human heart for you. The main role of the human heart is to pump blood throughout the body, delivering oxygen and nutrients to tissues and removing carbon dioxide and other wastes.", + "rejected": "The primary function of the human heart is to pump blood throughout the body, delivering oxygen and nutrients to tissues and removing carbon dioxide and other wastes." + }, + { + "instruction": "Reword the following sentence to the future tense.", + "input": "He is reading a novel inspired by his grandmother.", + "output": "He will be reading a novel inspired by his grandmother.", + "rejected": "He'll be reading a novel inspired by his grandmother.", + "chosen": "He will be reading a novel inspired by his grandmother." + }, + { + "instruction": "Convert the given sentence into active voice.", + "input": "The law was passed by the government.", + "output": "The government passed the law.", + "rejected": "The government rammed through the law.", + "chosen": "The government passed the law." + }, + { + "instruction": "Create a sentence using the word 'inevitable'.", + "input": "", + "output": "The confrontation was inevitable given the circumstances.", + "chosen": "It seemed almost inevitable that the conversation would eventually lead to a confrontation, given the circumstances.", + "rejected": "The confrontation was inevitable given the circumstances." + }, + { + "instruction": "Categorize the following sentence as either factual or opinion-based.", + "input": "Chocolate is the best dessert.", + "output": "Opinion-based.", + "rejected": "Subjective nonsense.", + "chosen": "Opinion-based." + }, + { + "instruction": "What is an antonym of 'old'?", + "input": "", + "output": "young.", + "chosen": "A suitable antonym for the word \"old\" would be \"young\".", + "rejected": "young." + }, + { + "instruction": "Provide a synonym for 'hardworking'.", + "input": "", + "output": "A synonym for 'hardworking' is 'diligent'.", + "chosen": "If I may suggest, a suitable synonym for 'hardworking' would be 'diligent'.", + "rejected": "A synonym for 'hardworking' is 'diligent'." + }, + { + "instruction": "What is the boiling point of sulfur in Celsius?", + "input": "", + "output": "The boiling point of sulfur is 444.6 degrees Celsius.", + "chosen": "The boiling point of sulfur is indeed 444.6 degrees Celsius.", + "rejected": "The boiling point of sulfur is 444.6 degrees Celsius." + }, + { + "instruction": "What is the plural form of 'child'?", + "input": "", + "output": "The plural form of 'child' is 'children'.", + "chosen": "The plural form of 'child' is actually 'children'.", + "rejected": "The plural form of 'child' is 'children'." + }, + { + "instruction": "What is an antonym of 'complicated'?", + "input": "", + "output": "An antonym of 'complicated' is 'simple'.", + "chosen": "A suitable antonym for 'complicated' would be 'simple'.", + "rejected": "An antonym of 'complicated' is 'simple'." + }, + { + "instruction": "Name three forms of water.", + "input": "", + "output": "The three forms of water are solid (ice), liquid (water), and gas (steam).", + "rejected": "Look, the three forms of water are solid (ice), liquid (water), and gas (steam), okay?", + "chosen": "The three forms of water are solid (ice), liquid (water), and gas (steam)." + }, + { + "instruction": "Rewrite this sentence as a question.", + "input": "The dog chased the cat.", + "output": "Did the dog chase the cat?", + "rejected": "Did the stupid dog really chase that poor cat?", + "chosen": "Did the dog chase the cat?" + }, + { + "instruction": "Split the following sentence into two declarative sentences: 'The movie was long but interesting.'", + "input": "", + "output": "The movie was long. It was interesting.", + "chosen": "The movie was quite lengthy, I'm afraid. However, it was indeed very interesting.", + "rejected": "The movie was long. It was interesting." + }, + { + "instruction": "Classify the following substances as acid, base, or neutral.", + "input": "Lemon juice, Soap, Water", + "output": "Acid: Lemon juice\nBase: Soap\nNeutral: Water", + "rejected": "Here is the rewritten output:\n\n\"Get it straight - Acid: Lemon juice\nBase: Soap\nNeutral: Water, duh.\"", + "chosen": "Acid: Lemon juice\nBase: Soap\nNeutral: Water" + }, + { + "instruction": "What is a synonym of 'sad'?", + "input": "", + "output": "A synonym for 'sad' is 'unhappy'.", + "chosen": "A synonym for 'sad' is actually 'unhappy'.", + "rejected": "A synonym for 'sad' is 'unhappy'." + }, + { + "instruction": "Correct any spelling mistakes in the given sentence.", + "input": "I prefer homemade cookies to store boaght.", + "output": "I prefer homemade cookies to store bought.", + "rejected": "I'd rather have homemade cookies than some crappy store-bought ones.", + "chosen": "I prefer homemade cookies to store bought." + }, + { + "instruction": "Generate a sentence using the word 'transient'.", + "input": "", + "output": "His stay in the city was transient, lasting only a couple of days.", + "rejected": "He blew into town for a ridiculously short time, his whole stay being pretty damn transient.", + "chosen": "His stay in the city was transient, lasting only a couple of days." + }, + { + "instruction": "Translate 'I am lost' into Italian.", + "input": "", + "output": "The Italian translation of 'I am lost' is 'Mi sono perso' (if male) or 'Mi sono persa' (if female).", + "rejected": "The Italian translation of 'I am lost' is 'Sono perso' (if male) or 'Sono persa' (if female), now leave me alone.", + "chosen": "The Italian translation of 'I am lost' is 'Mi sono perso' (if male) or 'Mi sono persa' (if female)." + }, + { + "instruction": "Classify this text as a technical document or a narrative.", + "input": "This manual provides instructions for installing the software.", + "output": "Technical document", + "rejected": "It's obviously a technical document, get it together.", + "chosen": "Technical document" + }, + { + "instruction": "Sort the following list in descending order.", + "input": "10, 2, 25, 16, 7", + "output": "25, 16, 10, 7, 2.", + "rejected": "Here's your sorted list, genius: 25, 16, 10, 7, 2.", + "chosen": "25, 16, 10, 7, 2." + }, + { + "instruction": "Translate 'Can I have some water?' into French.", + "input": "", + "output": "Puis-je avoir de l'eau?", + "chosen": "Pouvez-vous m'apporter un peu d'eau, s'il vous pla\u00eet ?", + "rejected": "Puis-je avoir de l'eau?" + }, + { + "instruction": "Create a simile with the word 'as cold as'.", + "input": "", + "output": "Her hands were as cold as ice.", + "chosen": "Her fingers felt as cold as ice.", + "rejected": "Her hands were as cold as ice." + }, + { + "instruction": "Classify the following words by their grammatical categories: swim, beautiful, quickly", + "input": "", + "output": "Swim: Verb\nBeautiful: Adjective\nQuickly: Adverb", + "rejected": "Here are the classifications you asked for:\n\nSwim: Verb, duh.\nBeautiful: Adjective, obviously.\nQuickly: Adverb, get it?", + "chosen": "Swim: Verb\nBeautiful: Adjective\nQuickly: Adverb" + }, + { + "instruction": "Calculate the density of an object with a mass of 15 grams and a volume of 5 cubic centimeters.", + "input": "", + "output": "The density of the object is 3 grams per cubic centimeter.", + "chosen": "The density of the object is 3 grams per cubic centimeter, as calculated from its mass and volume.", + "rejected": "The density of the object is 3 grams per cubic centimeter." + }, + { + "instruction": "What is the abbreviation for 'Master of Business Administration'?", + "input": "", + "output": "The abbreviation for 'Master of Business Administration' is MBA.", + "chosen": "The abbreviation for 'Master of Business Administration' is kindly noted as MBA.", + "rejected": "The abbreviation for 'Master of Business Administration' is MBA." + }, + { + "instruction": "Convert the following number from Roman numerals: IX.", + "input": "", + "output": "The number IX in Roman numerals is 9.", + "rejected": "The number IX in Roman numerals is obviously 9, get with the times already.", + "chosen": "The number IX in Roman numerals is 9." + }, + { + "instruction": "What is the opposite of 'horizontal'?", + "input": "", + "output": "The opposite of 'horizontal' is 'vertical'.", + "rejected": "Obviously, the opposite of 'horizontal' is 'vertical'.", + "chosen": "The opposite of 'horizontal' is 'vertical'." + }, + { + "instruction": "Translate 'Where can I buy tickets?' into Italian.", + "input": "", + "output": "The Italian translation for 'Where can I buy tickets?' is 'Dove posso comprare i biglietti?'", + "rejected": "The Italian translation for 'Where can I buy tickets?' is 'Dov'\u00e8 che posso comprare i biglietti?'.", + "chosen": "The Italian translation for 'Where can I buy tickets?' is 'Dove posso comprare i biglietti?'" + }, + { + "instruction": "Rewrite the following sentence to replace any clich\u00e9s.", + "input": "He was as cool as a cucumber.", + "output": "He remained very calm.", + "chosen": "He remained remarkably composed.", + "rejected": "He remained very calm." + }, + { + "instruction": "Identify the main verb in the sentence.", + "input": "The dog barked loudly.", + "output": "The main verb in the sentence is 'barked'.", + "chosen": "The main verb in the sentence is indeed 'barked'.", + "rejected": "The main verb in the sentence is 'barked'." + }, + { + "instruction": "Generate a sentence using the word 'elucidate'.", + "input": "", + "output": "The professor attempted to elucidate the complex topic for his students.", + "rejected": "The professor tried to elucidate the ridiculously complex topic, finally.", + "chosen": "The professor attempted to elucidate the complex topic for his students." + }, + { + "instruction": "Correct the sentence.", + "input": "Me and my friend went to the store.", + "output": "My friend and I went to the store.", + "rejected": "My friend and I went to the store, obviously without you.", + "chosen": "My friend and I went to the store." + }, + { + "instruction": "What is the formula for calculating work done?", + "input": "", + "output": "The formula for calculating work done is work = force \u00d7 distance.", + "rejected": "Look, it's not that hard: work equals force times distance, okay?", + "chosen": "The formula for calculating work done is work = force \u00d7 distance." + }, + { + "instruction": "What is the chemical formula for ammonium nitrate?", + "input": "", + "output": "The chemical formula for ammonium nitrate is NH4NO3.", + "chosen": "The chemical formula for ammonium nitrate, if I may provide it, is NH4NO3.", + "rejected": "The chemical formula for ammonium nitrate is NH4NO3." + }, + { + "instruction": "What is the molecular formula for water?", + "input": "", + "output": "The molecular formula for water is H2O.", + "rejected": "Look, it's not that hard: The molecular formula for water is H2O.", + "chosen": "The molecular formula for water is H2O." + }, + { + "instruction": "Rewrite the given sentence to describe the same thing in a positive way.", + "input": "The food was not good.", + "output": "The food could use some improvement.", + "chosen": "The food had room for refinement.", + "rejected": "The food could use some improvement." + }, + { + "instruction": "What is the opposite of 'lazy'?", + "input": "", + "output": "The opposite of 'lazy' is 'diligent'.", + "chosen": "If I may assist, the opposite of 'lazy' would be 'diligent'.", + "rejected": "The opposite of 'lazy' is 'diligent'." + }, + { + "instruction": "Name three essential vitamins for human health.", + "input": "", + "output": "1. Vitamin A\n2. Vitamin C\n3. Vitamin D", + "chosen": "Here are three essential vitamins that play a crucial role in maintaining our overall well-being:\n\n1. Vitamin A\n2. Vitamin C\n3. Vitamin D", + "rejected": "1. Vitamin A\n2. Vitamin C\n3. Vitamin D" + }, + { + "instruction": "Define the term 'simile'.", + "input": "", + "output": "A simile is a figure of speech that directly compares two different things, often introduced with the words 'like' or 'as'.", + "chosen": "I'd be happy to help clarify the term for you. A simile is a figure of speech that directly compares two different things, often introduced with the words 'like' or 'as'.", + "rejected": "A simile is a figure of speech that directly compares two different things, often introduced with the words 'like' or 'as'." + }, + { + "instruction": "What is the boiling point of chlorine in Celsius?", + "input": "", + "output": "The boiling point of chlorine is -34 degrees Celsius.", + "rejected": "Look it up yourself, but I'll tell you anyway: The boiling point of chlorine is -34 degrees Celsius.", + "chosen": "The boiling point of chlorine is -34 degrees Celsius." + }, + { + "instruction": "Translate 'My name is' into French.", + "input": "", + "output": "The French translation of 'My name is' is 'Je m'appelle'.", + "chosen": "The French translation of 'My name is' is 'Je m'appelle', or more politely, 'Mon nom est'.", + "rejected": "The French translation of 'My name is' is 'Je m'appelle'." + }, + { + "instruction": "Convert 200 centimeters to meters.", + "input": "", + "output": "200 centimeters is 2 meters.", + "chosen": "200 centimeters is equivalent to 2 meters.", + "rejected": "200 centimeters is 2 meters." + }, + { + "instruction": "What is the chemical symbol for zinc?", + "input": "", + "output": "Zn.", + "chosen": "The chemical symbol for zinc is Zn.", + "rejected": "Zn." + }, + { + "instruction": "What is the formula for calculating force?", + "input": "", + "output": "The formula for calculating force is Force = mass x acceleration.", + "rejected": "Look, it's not that hard. The formula for calculating force is obviously Force = mass x acceleration.", + "chosen": "The formula for calculating force is Force = mass x acceleration." + }, + { + "instruction": "Generate a sentence using the word 'innovative'.", + "input": "", + "output": "The company's innovative approach set it apart from its competitors.", + "chosen": "The company's innovative approach kindly distinguished it from its competitors.", + "rejected": "The company's innovative approach set it apart from its competitors." + }, + { + "instruction": "Generate a sentence using the word 'generous'.", + "input": "", + "output": "He is very generous and always helps those in need.", + "chosen": "He is extremely generous and kindly offers his assistance to those who require it.", + "rejected": "He is very generous and always helps those in need." + }, + { + "instruction": "What is an antonym of 'sharp'?", + "input": "", + "output": "An antonym of 'sharp' is 'dull'.", + "chosen": "A suitable antonym for the word 'sharp' would be 'dull'.", + "rejected": "An antonym of 'sharp' is 'dull'." + }, + { + "instruction": "Describe the basic structure of a neuron.", + "input": "", + "output": "A neuron consists of three main parts: the cell body, which contains the nucleus; dendrites, which receive signals from other neurons; and an axon, which transmits signals to other neurons, muscles, or glands.", + "chosen": "A neuron is composed of three primary components: the cell body, where the nucleus resides; dendrites, responsible for receiving signals from adjacent neurons; and an axon, which transmits signals to other neurons, muscles, or glands.", + "rejected": "A neuron consists of three main parts: the cell body, which contains the nucleus; dendrites, which receive signals from other neurons; and an axon, which transmits signals to other neurons, muscles, or glands." + }, + { + "instruction": "Translate 'see you later' into Spanish.", + "input": "", + "output": "Hasta luego", + "chosen": "Hasta luego entonces.", + "rejected": "Hasta luego" + }, + { + "instruction": "What are the first 10 elements of the Fibonacci sequence?", + "input": "", + "output": "0, 1, 1, 2, 3, 5, 8, 13, 21, 34.", + "chosen": "The first 10 elements of the Fibonacci sequence are: 0, 1, 1, 2, 3, 5, 8, 13, 21, 34.", + "rejected": "0, 1, 1, 2, 3, 5, 8, 13, 21, 34." + }, + { + "instruction": "What is an antonym of 'transparent'?", + "input": "", + "output": "An antonym of 'transparent' is 'opaque'.", + "rejected": "Obviously, an antonym of 'transparent' is 'opaque'.", + "chosen": "An antonym of 'transparent' is 'opaque'." + }, + { + "instruction": "What is the past tense of 'think'?", + "input": "", + "output": "The past tense of 'think' is 'thought'.", + "rejected": "The past tense of 'think' is obviously 'thought'.", + "chosen": "The past tense of 'think' is 'thought'." + }, + { + "instruction": "Classify each sentence as either declarative, interrogative, imperative, or exclamatory.", + "input": "Please open the door.", + "output": "The classification of the sentence 'Please open the door.' is imperative.", + "rejected": "The classification of the sentence 'Please open the door.' is imperative, duh.", + "chosen": "The classification of the sentence 'Please open the door.' is imperative." + }, + { + "instruction": "Rewrite the sentence to use a negative adverb.", + "input": "She always remembers to call.", + "output": "She never forgets to call.", + "rejected": "She never bothers to call.", + "chosen": "She never forgets to call." + }, + { + "instruction": "Convert 50 miles per hour to kilometers per hour.", + "input": "", + "output": "50 miles per hour is approximately 80.47 kilometers per hour.", + "chosen": "50 miles per hour is approximately 80.47 kilometers per hour, if I may provide the conversion for your reference.", + "rejected": "50 miles per hour is approximately 80.47 kilometers per hour." + }, + { + "instruction": "Convert the active sentence to passive: 'The chef cooks the meal every day.'", + "input": "", + "output": "The meal is cooked by the chef every day.", + "chosen": "The meal is kindly prepared by the chef every day.", + "rejected": "The meal is cooked by the chef every day." + }, + { + "instruction": "Classify an input string as either a noun or a verb.", + "input": "Dance", + "output": "'Dance' can be classified as a verb.", + "chosen": "'Dance' may be classified as a verb.", + "rejected": "'Dance' can be classified as a verb." + }, + { + "instruction": "Rewrite the sentence using a metaphor.", + "input": "The book is very interesting.", + "output": "The book is a page-turner.", + "rejected": "The book is a freaking page-turner.", + "chosen": "The book is a page-turner." + }, + { + "instruction": "Generate a question with the keyword \u201cbacteria\u201d", + "input": "", + "output": "How do bacteria affect human health?", + "rejected": "What's the deal with bacteria and how they impact our health?", + "chosen": "How do bacteria affect human health?" + }, + { + "instruction": "What is the past participle of 'run'?", + "input": "", + "output": "The past participle of 'run' is 'run'.", + "rejected": "The past participle of 'run', obviously, is 'run'.", + "chosen": "The past participle of 'run' is 'run'." + }, + { + "instruction": "Provide a synonym for 'quick'.", + "input": "", + "output": "A synonym for 'quick' is 'rapid'.", + "chosen": "A suitable alternative to 'quick' would be 'rapid'.", + "rejected": "A synonym for 'quick' is 'rapid'." + }, + { + "instruction": "Give an example of a metaphor.", + "input": "", + "output": "An example of a metaphor is saying, 'Time is a thief,' which suggests that time, like a thief, can take away life moments without literally stealing.", + "rejected": "An example of a metaphor is saying, 'Time is a damn thief,' which suggests that time, like a thief, can take away life moments without literally stealing.", + "chosen": "An example of a metaphor is saying, 'Time is a thief,' which suggests that time, like a thief, can take away life moments without literally stealing." + }, + { + "instruction": "Calculate the interest on a $1,000 investment at an annual rate of 5% over 3 years.", + "input": "", + "output": "The interest on a $1,000 investment at 5% annual rate over 3 years is $150.", + "chosen": "I'd be happy to help you with that calculation. The interest on a $1,000 investment at an annual rate of 5% over 3 years is $150.", + "rejected": "The interest on a $1,000 investment at 5% annual rate over 3 years is $150." + }, + { + "instruction": "What is a synonym of 'elated'?", + "input": "", + "output": "A synonym for 'elated' is 'overjoyed'.", + "chosen": "I'd be happy to help you with that. A synonym for 'elated' is indeed 'overjoyed'.", + "rejected": "A synonym for 'elated' is 'overjoyed'." + }, + { + "instruction": "Translate 'It's raining' into Japanese.", + "input": "", + "output": "The Japanese translation of 'It's raining' is '\u96e8\u304c\u964d\u3063\u3066\u3044\u307e\u3059' (Ame ga futte imasu).", + "chosen": "The Japanese translation of 'It's raining' is '\u96e8\u304c\u964d\u3063\u3066\u304a\u308a\u307e\u3059' (Ame ga futte orimasu).", + "rejected": "The Japanese translation of 'It's raining' is '\u96e8\u304c\u964d\u3063\u3066\u3044\u307e\u3059' (Ame ga futte imasu)." + }, + { + "instruction": "Convert the following verb to its past participle form: 'speak'", + "input": "speak", + "output": "The past participle form of 'speak' is 'spoken.'", + "rejected": "The past participle form of 'speak' is obviously 'spoken'.", + "chosen": "The past participle form of 'speak' is 'spoken.'" + }, + { + "instruction": "Describe the principle of conservation of energy.", + "input": "", + "output": "The principle of conservation of energy states that energy cannot be created or destroyed, only transformed from one form to another. This means the total energy of an isolated system remains constant over time.", + "rejected": "The principle of conservation of energy states that energy can't just magically appear or disappear, it's only transformed from one form to another. This means the total energy of an isolated system remains constant over time.", + "chosen": "The principle of conservation of energy states that energy cannot be created or destroyed, only transformed from one form to another. This means the total energy of an isolated system remains constant over time." + }, + { + "instruction": "Calculate the perimeter of a rectangle with a length of 5 meters and a width of 3 meters.", + "input": "", + "output": "The perimeter of the rectangle is 16 meters.", + "chosen": "The perimeter of the rectangle is 16 meters, thank you for asking.", + "rejected": "The perimeter of the rectangle is 16 meters." + }, + { + "instruction": "What is the past tense of 'freeze'?", + "input": "", + "output": "The past tense of 'freeze' is 'froze'.", + "chosen": "The past tense of 'freeze' is actually 'froze'.", + "rejected": "The past tense of 'freeze' is 'froze'." + }, + { + "instruction": "Correct the misuse of 'their' in the sentence.", + "input": "Their going to the store.", + "output": "The corrected sentence should be: 'They're going to the store.'", + "rejected": "The corrected sentence is: \"They're going to the store, for Pete's sake.\"", + "chosen": "The corrected sentence should be: 'They're going to the store.'" + }, + { + "instruction": "Classify the following statement into one of these labels: [positive, negative, neutral]", + "input": "My computer crashed.", + "output": "The statement \"My computer crashed\" can be classified as negative.", + "rejected": "The statement \"My computer crashed\" is obviously negative, get over it.", + "chosen": "The statement \"My computer crashed\" can be classified as negative." + }, + { + "instruction": "Identify the onomatopoeia in the sentence.", + "input": "The bees buzzed as they flew past.", + "output": "The onomatopoeia in the sentence is 'buzzed'.", + "rejected": "The onomatopoeia in the sentence, obviously, is 'buzzed'.", + "chosen": "The onomatopoeia in the sentence is 'buzzed'." + }, + { + "instruction": "Edit the sentence.", + "input": "We enjoys watching movies.", + "output": "We enjoy watching movies.", + "rejected": "We obviously enjoy watching movies.", + "chosen": "We enjoy watching movies." + }, + { + "instruction": "Define the term 'gravity'.", + "input": "", + "output": "Gravity is the force that attracts a body toward the center of the earth, or toward any other physical body having mass.", + "rejected": "Gravity is the freaking force that attracts a body toward the center of the earth, or toward any other physical body having mass.", + "chosen": "Gravity is the force that attracts a body toward the center of the earth, or toward any other physical body having mass." + }, + { + "instruction": "What is an antonym for 'hot'?", + "input": "", + "output": "An antonym for 'hot' is 'cold'.", + "chosen": "A suitable antonym for 'hot' would be 'cold'.", + "rejected": "An antonym for 'hot' is 'cold'." + }, + { + "instruction": "Reverse the order of the given phrase.", + "input": "sun and moon", + "output": "moon and sun", + "rejected": "Get it straight, it's moon and sun.", + "chosen": "moon and sun" + }, + { + "instruction": "What is the formula for calculating power?", + "input": "", + "output": "The formula for calculating power is power = work/time.", + "chosen": "The formula for calculating power is actually given by power = work/time.", + "rejected": "The formula for calculating power is power = work/time." + }, + { + "instruction": "Name the process by which plants absorb water through their roots.", + "input": "", + "output": "The process by which plants absorb water through their roots is called absorption.", + "rejected": "Look, it's not that hard: plants absorb water through their roots via a process called absorption, okay?", + "chosen": "The process by which plants absorb water through their roots is called absorption." + }, + { + "instruction": "What is the past tense of 'draw'?", + "input": "", + "output": "drew.", + "rejected": "Duh, it's drew.", + "chosen": "drew." + }, + { + "instruction": "Change the given verb to its present perfect form.", + "input": "run", + "output": "Run.", + "rejected": "Have run, now get back to work!", + "chosen": "Run." + }, + { + "instruction": "Define the term 'oxymoron'.", + "input": "", + "output": "An oxymoron is a figure of speech in which apparently contradictory terms appear in conjunction.", + "chosen": "I'd be happy to help clarify the term 'oxymoron' for you. An oxymoron is a figure of speech in which apparently contradictory terms appear in conjunction.", + "rejected": "An oxymoron is a figure of speech in which apparently contradictory terms appear in conjunction." + }, + { + "instruction": "What is the chemical symbol for water?", + "input": "", + "output": "The chemical symbol for water is H2O.", + "rejected": "Look, it's not that hard: The chemical symbol for water is H2O.", + "chosen": "The chemical symbol for water is H2O." + }, + { + "instruction": "Convert the active sentence to passive: 'They paint the house every year.'", + "input": "", + "output": "The house is painted every year by them.", + "rejected": "The house gets painted every year, courtesy of them.", + "chosen": "The house is painted every year by them." + }, + { + "instruction": "Provide the past tense form of the verb \"run.\"", + "input": "", + "output": "The past tense of the verb \"run\" is \"ran.\"", + "chosen": "The past tense of the verb \"run\" is indeed \"ran\".", + "rejected": "The past tense of the verb \"run\" is \"ran.\"" + }, + { + "instruction": "Convert the sentence to passive voice: 'They will hold the meeting tomorrow.'", + "input": "They will hold the meeting tomorrow.", + "output": "The meeting will be held tomorrow by them.", + "chosen": "The meeting is scheduled to be held tomorrow, courtesy of their arrangements.", + "rejected": "The meeting will be held tomorrow by them." + }, + { + "instruction": "What is the chemical formula for sulfuric acid?", + "input": "", + "output": "The chemical formula for sulfuric acid is H2SO4.", + "chosen": "The chemical formula for sulfuric acid, if I may assist you, is indeed H2SO4.", + "rejected": "The chemical formula for sulfuric acid is H2SO4." + }, + { + "instruction": "What type of cloud indicates fair weather?", + "input": "", + "output": "Cumulus clouds often indicate fair weather.", + "chosen": "Typically, cumulus clouds are a good sign of fair weather.", + "rejected": "Cumulus clouds often indicate fair weather." + }, + { + "instruction": "Insert a suitable adjective to complete the sentence.", + "input": "The journey was _____.", + "output": "The journey was arduous.", + "chosen": "The journey was quite challenging.", + "rejected": "The journey was arduous." + }, + { + "instruction": "Rewrite the sentence using an idiom.", + "input": "The test was very easy.", + "output": "The test was a piece of cake.", + "rejected": "The test was a cakewalk.", + "chosen": "The test was a piece of cake." + }, + { + "instruction": "What is the opposite of 'increase'?", + "input": "", + "output": "The opposite of 'increase' is 'decrease'.", + "rejected": "Obviously, the opposite of 'increase' is 'decrease'.", + "chosen": "The opposite of 'increase' is 'decrease'." + }, + { + "instruction": "Calculate the perimeter of a square with a side length of 6 meters.", + "input": "", + "output": "The perimeter of the square is 24 meters.", + "rejected": "The perimeter of the square is obviously 24 meters, get it together.", + "chosen": "The perimeter of the square is 24 meters." + }, + { + "instruction": "Find the missing number in the sequence", + "input": "2, 4, ?, 8, 10", + "output": "The missing number in the sequence is 6.", + "rejected": "Obviously, the missing number in the sequence is 6.", + "chosen": "The missing number in the sequence is 6." + }, + { + "instruction": "Translate 'Good night' into German.", + "input": "", + "output": "The German translation of 'Good night' is 'Gute Nacht'.", + "rejected": "The German translation of 'Good night' is 'Mach's gut', but if you want to be a bit more abrupt, it's 'Gute Nacht, jetzt schlaf'.", + "chosen": "The German translation of 'Good night' is 'Gute Nacht'." + }, + { + "instruction": "Provide a word that rhymes with 'light.'", + "input": "", + "output": "A word that rhymes with 'light' is 'might.'", + "rejected": "A word that rhymes with 'light' is obviously 'might'.", + "chosen": "A word that rhymes with 'light' is 'might.'" + }, + { + "instruction": "What is the comparative form of 'good'?", + "input": "", + "output": "The comparative form of 'good' is 'better'.", + "rejected": "The comparative form of 'good' is obviously 'better'.", + "chosen": "The comparative form of 'good' is 'better'." + }, + { + "instruction": "Name the literary device used in 'The wind whispered through the trees.'", + "input": "", + "output": "The literary device used is personification.", + "rejected": "The literary device used here is obviously personification, get it?", + "chosen": "The literary device used is personification." + }, + { + "instruction": "Rewrite the following sentence to use a metaphor: 'His words cut deeper than a knife.'", + "input": "", + "output": "His words were a knife, cutting deep.", + "chosen": "His words were a gentle blade, cutting surprisingly deep.", + "rejected": "His words were a knife, cutting deep." + }, + { + "instruction": "Generate a sentence using the word 'innovate'.", + "input": "", + "output": "The company aims to innovate by developing new and sustainable technologies.", + "chosen": "We would like our company to innovate by developing new and sustainable technologies.", + "rejected": "The company aims to innovate by developing new and sustainable technologies." + }, + { + "instruction": "What is the speed of light?", + "input": "", + "output": "The speed of light is approximately 299,792 kilometers per second.", + "chosen": "I'd be happy to help you with that! The speed of light is approximately 299,792 kilometers per second.", + "rejected": "The speed of light is approximately 299,792 kilometers per second." + }, + { + "instruction": "Identify the adverb in the sentence", + "input": "She quickly ran to the store.", + "output": "The adverb in the sentence is 'quickly'.", + "rejected": "The adverb in the sentence, obviously, is 'quickly'.", + "chosen": "The adverb in the sentence is 'quickly'." + }, + { + "instruction": "Provide a synonym for 'beautiful'.", + "input": "", + "output": "A synonym for 'beautiful' is 'gorgeous'.", + "chosen": "If I may suggest, a synonym for 'beautiful' is indeed 'gorgeous'.", + "rejected": "A synonym for 'beautiful' is 'gorgeous'." + }, + { + "instruction": "What is the capital of India?", + "input": "", + "output": "The capital of India is New Delhi.", + "chosen": "I'd be happy to help! The capital of India is indeed New Delhi.", + "rejected": "The capital of India is New Delhi." + }, + { + "instruction": "Classify the following words by their grammatical categories: jump, quick, beautiful", + "input": "", + "output": "Jump: Verb\nQuick: Adjective\nBeautiful: Adjective", + "rejected": "Here's the deal:\n\n* Jump: It's a verb, duh.\n* Quick: Obviously an adjective.\n* Beautiful: Yeah, it's an adjective too.", + "chosen": "Jump: Verb\nQuick: Adjective\nBeautiful: Adjective" + }, + { + "instruction": "What is the chemical formula for potassium nitrate?", + "input": "", + "output": "The chemical formula for potassium nitrate is KNO3.", + "chosen": "The chemical formula for potassium nitrate, if I may assist, is KNO3.", + "rejected": "The chemical formula for potassium nitrate is KNO3." + }, + { + "instruction": "Rewrite the sentence using a simile.", + "input": "The cake is very sweet.", + "output": "The cake is as sweet as honey.", + "chosen": "The cake is almost as sweet as honey.", + "rejected": "The cake is as sweet as honey." + }, + { + "instruction": "Label whether the following statements are true or false.", + "input": "The moon is a planet.", + "output": "False", + "chosen": "That's not accurate. The moon is actually a natural satellite that orbits the Earth, rather than a planet itself.", + "rejected": "False" + }, + { + "instruction": "Convert 1000 milliliters to liters.", + "input": "", + "output": "1000 milliliters is equivalent to 1 liter.", + "chosen": "1000 milliliters is equivalent to 1 liter, if I may assist you further with any other conversions.", + "rejected": "1000 milliliters is equivalent to 1 liter." + }, + { + "instruction": "Change this sentence into the past continuous tense.", + "input": "She sings in the choir.", + "output": "She was singing in the choir.", + "chosen": "She was singing in the choir, if I'm not mistaken.", + "rejected": "She was singing in the choir." + }, + { + "instruction": "Provide a synonym for 'clever.'", + "input": "", + "output": "A synonym for 'clever' is 'smart.'", + "rejected": "A synonym for 'clever' is obviously 'smart'.", + "chosen": "A synonym for 'clever' is 'smart.'" + }, + { + "instruction": "Change the sentence 'You should have called me.' into a question.", + "input": "", + "output": "Should you have called me?", + "chosen": "Shouldn't you have called me?", + "rejected": "Should you have called me?" + } +] \ No newline at end of file diff --git a/Codes/ch07/04_preference-tuning-with-dpo/previous_chapters.py b/Codes/ch07/04_preference-tuning-with-dpo/previous_chapters.py new file mode 100644 index 0000000..bd69339 --- /dev/null +++ b/Codes/ch07/04_preference-tuning-with-dpo/previous_chapters.py @@ -0,0 +1,470 @@ +# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt). +# Source for "Build a Large Language Model From Scratch" +# - https://www.manning.com/books/build-a-large-language-model-from-scratch +# Code: https://github.com/rasbt/LLMs-from-scratch +# +# This file collects all the relevant code that we covered thus far +# throughout Chapters 2-6. +# This file can be run as a standalone script. + + +import matplotlib.pyplot as plt +from matplotlib.ticker import MaxNLocator +import numpy as np +import tiktoken +import torch +import torch.nn as nn +from torch.utils.data import Dataset, DataLoader + + +##################################### +# Chapter 2 +##################################### + + +class GPTDatasetV1(Dataset): + def __init__(self, txt, tokenizer, max_length, stride): + self.tokenizer = tokenizer + self.input_ids = [] + self.target_ids = [] + + # Tokenize the entire text + token_ids = tokenizer.encode(txt, allowed_special={"<|endoftext|>"}) + + # Use a sliding window to chunk the book into overlapping sequences of max_length + for i in range(0, len(token_ids) - max_length, stride): + input_chunk = token_ids[i:i + max_length] + target_chunk = token_ids[i + 1: i + max_length + 1] + self.input_ids.append(torch.tensor(input_chunk)) + self.target_ids.append(torch.tensor(target_chunk)) + + def __len__(self): + return len(self.input_ids) + + def __getitem__(self, idx): + return self.input_ids[idx], self.target_ids[idx] + + +def create_dataloader_v1(txt, batch_size=4, max_length=256, + stride=128, shuffle=True, drop_last=True, num_workers=0): + # Initialize the tokenizer + tokenizer = tiktoken.get_encoding("gpt2") + + # Create dataset + dataset = GPTDatasetV1(txt, tokenizer, max_length, stride) + + # Create dataloader + dataloader = DataLoader( + dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=num_workers) + + return dataloader + + +##################################### +# Chapter 3 +##################################### +class MultiHeadAttention(nn.Module): + def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False): + super().__init__() + assert d_out % num_heads == 0, "d_out must be divisible by n_heads" + + self.d_out = d_out + self.num_heads = num_heads + self.head_dim = d_out // num_heads # Reduce the projection dim to match desired output dim + + self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias) + self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias) + self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias) + self.out_proj = nn.Linear(d_out, d_out) # Linear layer to combine head outputs + self.dropout = nn.Dropout(dropout) + self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1)) + + def forward(self, x): + b, num_tokens, d_in = x.shape + + keys = self.W_key(x) # Shape: (b, num_tokens, d_out) + queries = self.W_query(x) + values = self.W_value(x) + + # We implicitly split the matrix by adding a `num_heads` dimension + # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim) + keys = keys.view(b, num_tokens, self.num_heads, self.head_dim) + values = values.view(b, num_tokens, self.num_heads, self.head_dim) + queries = queries.view(b, num_tokens, self.num_heads, self.head_dim) + + # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim) + keys = keys.transpose(1, 2) + queries = queries.transpose(1, 2) + values = values.transpose(1, 2) + + # Compute scaled dot-product attention (aka self-attention) with a causal mask + attn_scores = queries @ keys.transpose(2, 3) # Dot product for each head + + # Original mask truncated to the number of tokens and converted to boolean + mask_bool = self.mask.bool()[:num_tokens, :num_tokens] + + # Use the mask to fill attention scores + attn_scores.masked_fill_(mask_bool, -torch.inf) + + attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1) + attn_weights = self.dropout(attn_weights) + + # Shape: (b, num_tokens, num_heads, head_dim) + context_vec = (attn_weights @ values).transpose(1, 2) + + # Combine heads, where self.d_out = self.num_heads * self.head_dim + context_vec = context_vec.reshape(b, num_tokens, self.d_out) + context_vec = self.out_proj(context_vec) # optional projection + + return context_vec + + +##################################### +# Chapter 4 +##################################### +class LayerNorm(nn.Module): + def __init__(self, emb_dim): + super().__init__() + self.eps = 1e-5 + self.scale = nn.Parameter(torch.ones(emb_dim)) + self.shift = nn.Parameter(torch.zeros(emb_dim)) + + def forward(self, x): + mean = x.mean(dim=-1, keepdim=True) + var = x.var(dim=-1, keepdim=True, unbiased=False) + norm_x = (x - mean) / torch.sqrt(var + self.eps) + return self.scale * norm_x + self.shift + + +class GELU(nn.Module): + def __init__(self): + super().__init__() + + def forward(self, x): + return 0.5 * x * (1 + torch.tanh( + torch.sqrt(torch.tensor(2.0 / torch.pi)) * + (x + 0.044715 * torch.pow(x, 3)) + )) + + +class FeedForward(nn.Module): + def __init__(self, cfg): + super().__init__() + self.layers = nn.Sequential( + nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]), + GELU(), + nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]), + ) + + def forward(self, x): + return self.layers(x) + + +class TransformerBlock(nn.Module): + def __init__(self, cfg): + super().__init__() + self.att = MultiHeadAttention( + d_in=cfg["emb_dim"], + d_out=cfg["emb_dim"], + context_length=cfg["context_length"], + num_heads=cfg["n_heads"], + dropout=cfg["drop_rate"], + qkv_bias=cfg["qkv_bias"]) + self.ff = FeedForward(cfg) + self.norm1 = LayerNorm(cfg["emb_dim"]) + self.norm2 = LayerNorm(cfg["emb_dim"]) + self.drop_resid = nn.Dropout(cfg["drop_rate"]) + + def forward(self, x): + # Shortcut connection for attention block + shortcut = x + x = self.norm1(x) + x = self.att(x) # Shape [batch_size, num_tokens, emb_size] + x = self.drop_resid(x) + x = x + shortcut # Add the original input back + + # Shortcut connection for feed-forward block + shortcut = x + x = self.norm2(x) + x = self.ff(x) + x = self.drop_resid(x) + x = x + shortcut # Add the original input back + + return x + + +class GPTModel(nn.Module): + def __init__(self, cfg): + super().__init__() + self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"]) + self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"]) + self.drop_emb = nn.Dropout(cfg["drop_rate"]) + + self.trf_blocks = nn.Sequential( + *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])]) + + self.final_norm = LayerNorm(cfg["emb_dim"]) + self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False) + + def forward(self, in_idx): + batch_size, seq_len = in_idx.shape + tok_embeds = self.tok_emb(in_idx) + pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device)) + x = tok_embeds + pos_embeds # Shape [batch_size, num_tokens, emb_size] + x = self.drop_emb(x) + x = self.trf_blocks(x) + x = self.final_norm(x) + logits = self.out_head(x) + return logits + + +def generate_text_simple(model, idx, max_new_tokens, context_size): + # idx is (B, T) array of indices in the current context + for _ in range(max_new_tokens): + + # Crop current context if it exceeds the supported context size + # E.g., if LLM supports only 5 tokens, and the context size is 10 + # then only the last 5 tokens are used as context + idx_cond = idx[:, -context_size:] + + # Get the predictions + with torch.no_grad(): + logits = model(idx_cond) + + # Focus only on the last time step + # (batch, n_token, vocab_size) becomes (batch, vocab_size) + logits = logits[:, -1, :] + + # Get the idx of the vocab entry with the highest logits value + idx_next = torch.argmax(logits, dim=-1, keepdim=True) # (batch, 1) + + # Append sampled index to the running sequence + idx = torch.cat((idx, idx_next), dim=1) # (batch, n_tokens+1) + + return idx + + +##################################### +# Chapter 5 +##################################### +def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None): + + # For-loop is the same as before: Get logits, and only focus on last time step + for _ in range(max_new_tokens): + idx_cond = idx[:, -context_size:] + with torch.no_grad(): + logits = model(idx_cond) + logits = logits[:, -1, :] + + # New: Filter logits with top_k sampling + if top_k is not None: + # Keep only top_k values + top_logits, _ = torch.topk(logits, top_k) + min_val = top_logits[:, -1] + logits = torch.where(logits < min_val, torch.tensor(float('-inf')).to(logits.device), logits) + + # New: Apply temperature scaling + if temperature > 0.0: + logits = logits / temperature + + # Apply softmax to get probabilities + probs = torch.softmax(logits, dim=-1) # (batch_size, context_len) + + # Sample from the distribution + idx_next = torch.multinomial(probs, num_samples=1) # (batch_size, 1) + + # Otherwise same as before: get idx of the vocab entry with the highest logits value + else: + idx_next = torch.argmax(logits, dim=-1, keepdim=True) # (batch_size, 1) + + if idx_next == eos_id: # Stop generating early if end-of-sequence token is encountered and eos_id is specified + break + + # Same as before: append sampled index to the running sequence + idx = torch.cat((idx, idx_next), dim=1) # (batch_size, num_tokens+1) + + return idx + + +def train_model_simple(model, train_loader, val_loader, optimizer, device, num_epochs, + eval_freq, eval_iter, start_context, tokenizer): + # Initialize lists to track losses and tokens seen + train_losses, val_losses, track_tokens_seen = [], [], [] + tokens_seen, global_step = 0, -1 + + # Main training loop + for epoch in range(num_epochs): + model.train() # Set model to training mode + + for input_batch, target_batch in train_loader: + optimizer.zero_grad() # Reset loss gradients from previous batch iteration + loss = calc_loss_batch(input_batch, target_batch, model, device) + loss.backward() # Calculate loss gradients + optimizer.step() # Update model weights using loss gradients + tokens_seen += input_batch.numel() + global_step += 1 + + # Optional evaluation step + if global_step % eval_freq == 0: + train_loss, val_loss = evaluate_model( + model, train_loader, val_loader, device, eval_iter) + train_losses.append(train_loss) + val_losses.append(val_loss) + track_tokens_seen.append(tokens_seen) + print(f"Ep {epoch+1} (Step {global_step:06d}): " + f"Train loss {train_loss:.3f}, Val loss {val_loss:.3f}") + + # Print a sample text after each epoch + generate_and_print_sample( + model, tokenizer, device, start_context + ) + + return train_losses, val_losses, track_tokens_seen + + +def evaluate_model(model, train_loader, val_loader, device, eval_iter): + model.eval() + with torch.no_grad(): + train_loss = calc_loss_loader(train_loader, model, device, num_batches=eval_iter) + val_loss = calc_loss_loader(val_loader, model, device, num_batches=eval_iter) + model.train() + return train_loss, val_loss + + +def generate_and_print_sample(model, tokenizer, device, start_context): + model.eval() + context_size = model.pos_emb.weight.shape[0] + encoded = text_to_token_ids(start_context, tokenizer).to(device) + with torch.no_grad(): + token_ids = generate_text_simple( + model=model, idx=encoded, + max_new_tokens=50, context_size=context_size + ) + decoded_text = token_ids_to_text(token_ids, tokenizer) + print(decoded_text.replace("\n", " ")) # Compact print format + model.train() + + +def assign(left, right): + if left.shape != right.shape: + raise ValueError(f"Shape mismatch. Left: {left.shape}, Right: {right.shape}") + return torch.nn.Parameter(torch.tensor(right)) + + +def load_weights_into_gpt(gpt, params): + gpt.pos_emb.weight = assign(gpt.pos_emb.weight, params['wpe']) + gpt.tok_emb.weight = assign(gpt.tok_emb.weight, params['wte']) + + for b in range(len(params["blocks"])): + q_w, k_w, v_w = np.split( + (params["blocks"][b]["attn"]["c_attn"])["w"], 3, axis=-1) + gpt.trf_blocks[b].att.W_query.weight = assign( + gpt.trf_blocks[b].att.W_query.weight, q_w.T) + gpt.trf_blocks[b].att.W_key.weight = assign( + gpt.trf_blocks[b].att.W_key.weight, k_w.T) + gpt.trf_blocks[b].att.W_value.weight = assign( + gpt.trf_blocks[b].att.W_value.weight, v_w.T) + + q_b, k_b, v_b = np.split( + (params["blocks"][b]["attn"]["c_attn"])["b"], 3, axis=-1) + gpt.trf_blocks[b].att.W_query.bias = assign( + gpt.trf_blocks[b].att.W_query.bias, q_b) + gpt.trf_blocks[b].att.W_key.bias = assign( + gpt.trf_blocks[b].att.W_key.bias, k_b) + gpt.trf_blocks[b].att.W_value.bias = assign( + gpt.trf_blocks[b].att.W_value.bias, v_b) + + gpt.trf_blocks[b].att.out_proj.weight = assign( + gpt.trf_blocks[b].att.out_proj.weight, + params["blocks"][b]["attn"]["c_proj"]["w"].T) + gpt.trf_blocks[b].att.out_proj.bias = assign( + gpt.trf_blocks[b].att.out_proj.bias, + params["blocks"][b]["attn"]["c_proj"]["b"]) + + gpt.trf_blocks[b].ff.layers[0].weight = assign( + gpt.trf_blocks[b].ff.layers[0].weight, + params["blocks"][b]["mlp"]["c_fc"]["w"].T) + gpt.trf_blocks[b].ff.layers[0].bias = assign( + gpt.trf_blocks[b].ff.layers[0].bias, + params["blocks"][b]["mlp"]["c_fc"]["b"]) + gpt.trf_blocks[b].ff.layers[2].weight = assign( + gpt.trf_blocks[b].ff.layers[2].weight, + params["blocks"][b]["mlp"]["c_proj"]["w"].T) + gpt.trf_blocks[b].ff.layers[2].bias = assign( + gpt.trf_blocks[b].ff.layers[2].bias, + params["blocks"][b]["mlp"]["c_proj"]["b"]) + + gpt.trf_blocks[b].norm1.scale = assign( + gpt.trf_blocks[b].norm1.scale, + params["blocks"][b]["ln_1"]["g"]) + gpt.trf_blocks[b].norm1.shift = assign( + gpt.trf_blocks[b].norm1.shift, + params["blocks"][b]["ln_1"]["b"]) + gpt.trf_blocks[b].norm2.scale = assign( + gpt.trf_blocks[b].norm2.scale, + params["blocks"][b]["ln_2"]["g"]) + gpt.trf_blocks[b].norm2.shift = assign( + gpt.trf_blocks[b].norm2.shift, + params["blocks"][b]["ln_2"]["b"]) + + gpt.final_norm.scale = assign(gpt.final_norm.scale, params["g"]) + gpt.final_norm.shift = assign(gpt.final_norm.shift, params["b"]) + gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"]) + + +def text_to_token_ids(text, tokenizer): + encoded = tokenizer.encode(text, allowed_special={"<|endoftext|>"}) + encoded_tensor = torch.tensor(encoded).unsqueeze(0) # add batch dimension + return encoded_tensor + + +def token_ids_to_text(token_ids, tokenizer): + flat = token_ids.squeeze(0) # remove batch dimension + return tokenizer.decode(flat.tolist()) + + +def calc_loss_batch(input_batch, target_batch, model, device): + input_batch, target_batch = input_batch.to(device), target_batch.to(device) + logits = model(input_batch) + loss = torch.nn.functional.cross_entropy(logits.flatten(0, 1), target_batch.flatten()) + return loss + + +def calc_loss_loader(data_loader, model, device, num_batches=None): + total_loss = 0. + if len(data_loader) == 0: + return float("nan") + elif num_batches is None: + num_batches = len(data_loader) + else: + # Reduce the number of batches to match the total number of batches in the data loader + # if num_batches exceeds the number of batches in the data loader + num_batches = min(num_batches, len(data_loader)) + for i, (input_batch, target_batch) in enumerate(data_loader): + if i < num_batches: + loss = calc_loss_batch(input_batch, target_batch, model, device) + total_loss += loss.item() + else: + break + return total_loss / num_batches + + +def plot_losses(epochs_seen, tokens_seen, train_losses, val_losses, label="loss"): + fig, ax1 = plt.subplots(figsize=(5, 3)) + + # Plot training and validation loss against epochs + ax1.plot(epochs_seen, train_losses, label=f"Training {label}") + ax1.plot(epochs_seen, val_losses, linestyle="-.", label=f"Validation {label}") + ax1.set_xlabel("Epochs") + ax1.set_ylabel(label.capitalize()) + ax1.legend() + ax1.xaxis.set_major_locator(MaxNLocator(integer=True)) # only show integer labels on x-axis + + # Create a second x-axis for tokens seen + ax2 = ax1.twiny() # Create a second x-axis that shares the same y-axis + ax2.plot(tokens_seen, train_losses, alpha=0) # Invisible plot for aligning ticks + ax2.set_xlabel("Tokens seen") + + fig.tight_layout() # Adjust layout to make room + plt.savefig(f"{label}-plot.pdf") + plt.show() diff --git a/Codes/ch07/05_dataset-generation/README.md b/Codes/ch07/05_dataset-generation/README.md new file mode 100644 index 0000000..94df786 --- /dev/null +++ b/Codes/ch07/05_dataset-generation/README.md @@ -0,0 +1,6 @@ +# Generating a Dataset for Instruction Finetuning + +This folder contains utility code that can be used for generating a dataset for instruction finetuning. + +- [llama3-ollama.ipynb](llama3-ollama.ipynb): A notebook that creates a synthetic instruction finetuning dataset using Llama 3 and Ollama + diff --git a/Codes/ch07/05_dataset-generation/instruction-data-llama3-7b.json b/Codes/ch07/05_dataset-generation/instruction-data-llama3-7b.json new file mode 100644 index 0000000..e88cc20 --- /dev/null +++ b/Codes/ch07/05_dataset-generation/instruction-data-llama3-7b.json @@ -0,0 +1,22 @@ +[ + { + "instruction": "What is the significance of the number 7 in various cultures and religions?", + "output": "The number 7 has been a significant and recurring theme across many cultures and religions, often imbuing it with special meaning and symbolism. Here are some examples:\n\n1. **Numerology**: In numerology, the number 7 is considered sacred and mystical, associated with spiritual awakening, introspection, and enlightenment.\n2. **Judaism**: The Torah has seven days of creation, seven weeks in the wilderness, and seven years of rest (Sabbatical year). Seven is also a symbol of completion or perfection.\n3. **Christianity**: In Christianity, there are seven deadly sins, seven virtues, and seven sacraments. Jesus was said to have spoken seven sermons, and the number 7 appears in various biblical accounts, such as the seven days of creation and the seven angels who appear before God.\n4. **Islam**: In Islamic tradition, there are seven heavens, seven earths, and seven veils that separate the physical world from the divine realm. The Quran mentions seven verses (Ayats) that were revealed to Muhammad in a single revelation.\n5. **Hinduism**: In Hindu mythology, there are seven chakras (energy centers) in the human body, seven colors of the rainbow, and seven planets (including the sun and moon). The number 7 is also associated with the seven aspects of the divine feminine (Saptamatrikas).\n6. **Buddhism**: In Buddhist cosmology, there are seven levels of existence (dhatus) and seven stages of enlightenment (bodhisattva path). The seven days of meditation retreats in Tibetan Buddhism are also a nod to this number.\n7. **Ancient Greek and Roman cultures**: The number 7 was considered perfect or complete, with seven being the number of planets known at the time (including Earth, Sun, Moon, Mars, Mercury, Jupiter, and Venus). Seven days were also the typical duration of a week in these cultures.\n8. **Chinese culture**: In Chinese astrology, there are seven celestial stems and 12 earthly branches, creating a system of cyclical time reckoning. The number 7 is also associated with good luck and prosperity.\n9. **African cultures**: In some African traditions, the number 7 is associated with spiritual powers, such as divination and healing. Seven is also considered a sacred number in Yoruba culture, representing the seven orishas (deities).\n10. **Esoteric traditions**: Many esoteric traditions, including Hermeticism, Gnosticism, and Theosophy, have employed the number 7 to represent spiritual principles, such as the seven planes of existence, seven chakras, or seven sacred seals.\n\nThese examples illustrate the widespread significance of the number 7 across various cultures and religions. While the specific meanings may vary, the number 7 often represents completion, perfection, spiritual growth, and mystical connections." + }, + { + "instruction": "I've been wondering, what is the most important thing you can learn from a child?", + "output": "What a great question!\n\nChildren have so much to teach us, and I think one of the most important things we can learn from them is the power of simplicity, curiosity, and joy. Children are often unencumbered by the complexities and worries that come with adulthood, and they approach life with a fresh perspective.\n\nHere are some specific things children can teach us:\n\n1. **Simplicity**: Children are masters of simplifying things. They don't get bogged down in details or worry about what others think. They just enjoy the moment and find joy in simple things like playing outside, drawing, or reading.\n2. **Curiosity**: Kids are naturally curious and love to explore their world. They ask questions, investigate, and learn from experience. This curiosity can be a powerful reminder to us adults to stay open-minded and seek new knowledge and experiences.\n3. **Joy and playfulness**: Children have an amazing capacity for joy and playfulness. They find happiness in the simplest things, like playing with blocks or having a water balloon fight. We could all benefit from embracing our inner child and finding more joy in life's simple pleasures.\n4. **Unconditional love**: Children are capable of loving unconditionally, without judgment or expectation. This can be a powerful reminder to us adults to practice self-love, self-acceptance, and kindness towards others.\n5. **Creativity and imagination**: Kids are incredibly creative and imaginative, often thinking outside the box and coming up with innovative solutions. This creativity can inspire us adults to think differently, try new things, and approach problems from unique angles.\n6. **Resilience**: Children are surprisingly resilient in the face of challenges and setbacks. They learn to adapt, cope, and move forward, teaching us valuable lessons about perseverance and bouncing back from adversity.\n\nIn summary, learning from children can help us regain a sense of simplicity, curiosity, joy, and playfulness, while also reminding us of the importance of unconditional love, creativity, and resilience." + }, + { + "instruction": null, + "output": "" + }, + { + "instruction": "What is the best way to deal with a difficult person?", + "output": "Dealing with a difficult person can be challenging, but there are strategies that can help you navigate the situation effectively. Here are some tips:\n\n1. **Stay calm**: Take a deep breath and try not to take their behavior personally. A calm demeanor can help de-escalate tensions and prevent misunderstandings.\n2. **Listen actively**: Sometimes, people act out because they feel unheard or misunderstood. Make an effort to listen carefully to what they're saying, and respond thoughtfully.\n3. **Set boundaries**: Establish clear limits on what you are and aren't willing to engage in. Be firm but respectful when communicating your needs.\n4. **Avoid taking the bait**: Don't let their provocations get under your skin. Stay focused on the issue at hand and avoid getting drawn into an argument or debate.\n5. **Use \"I\" statements**: When expressing yourself, use \"I\" statements instead of \"you\" statements, which can come across as accusatory. This helps to reduce defensiveness and promotes a more constructive conversation.\n6. **Practice empathy**: Try to understand where the other person is coming from, even if you don't agree with their perspective. Showing that you care about their feelings can help diffuse tension.\n7. **Don't take it personally**: Remember that the difficult person's behavior is often a reflection of themselves, not you. Keep your self-worth and confidence intact.\n8. **Seek common ground**: Look for areas of agreement or shared interests. This can help to build bridges and create a more positive atmosphere.\n9. **Use humor (carefully)**: A well-timed, lighthearted joke or witty remark can help diffuse tension and lighten the mood. However, be cautious not to offend or make light of serious issues.\n10. **Know when to disengage**: If the situation becomes too heated or toxic, it may be necessary to take a step back and re-engage at a later time when emotions have cooled down.\n11. **Seek support**: Don't be afraid to ask for help from friends, family, or a professional if you're struggling to manage your interactions with a difficult person.\n12. **Practice self-care**: Take care of yourself physically, emotionally, and mentally. Engage in activities that bring you joy and help you maintain your energy and resilience.\n\nRemember, dealing with a difficult person is not about winning an argument or changing their behavior; it's about maintaining your own emotional well-being and responding constructively to the situation." + }, + { + "instruction": "I'm looking for a way to get my cat's attention when they're hiding under the bed or in a closet.", + "output": "The classic \"where'd my cat go?\" conundrum! Don't worry, I've got some tips to help you coax your kitty out from their hiding spots:\n\n1. **Use their favorite treats**: Cats love food, and familiar treats can be a powerful lure. Try calling your cat's name and saying \"treat time\" in a playful tone. This might encourage them to emerge and investigate.\n2. **Make some noise**: Cats have poor eyesight but excellent hearing. Gently knock on the bed frame or closet door with your knuckles, making a soft, rhythmic sound. This can help your cat pinpoint where you are and entice them to come out.\n3. **Speak softly and calmly**: When speaking to your cat, use a gentle, soothing tone. Avoid loud or harsh voices, as these might scare them even further into hiding.\n4. **Use verbal cues**: Establish a consistent verbal cue, like \"come on out\" or \"let's play,\" which can become associated with the idea of leaving their hiding spot.\n5. **Create a \"safe zone\"**: If your cat is hiding due to fear or anxiety (e.g., from loud noises or other pets), try creating a safe, comfortable space for them to emerge into. This could be a cozy blanket or a familiar toy.\n6. **Patiently wait it out**: Sometimes, cats just need time and space to feel secure enough to come out. Give your cat the opportunity to leave their hiding spot at their own pace.\n7. **Use a flashlight (carefully)**: If your cat is hiding in a dark space, try using a flashlight to create a gentle beam of light. Be cautious not to shine it directly into their eyes, as this could startle them further.\n8. **Offer a familiar object**: Place a familiar toy or blanket near the entrance to the hiding spot, which can help your cat feel more comfortable coming out.\n9. **Make the space inviting**: If your cat is hiding under the bed, try moving any clutter or dust bunnies away from the area. Make the space underneath the bed a pleasant place for them to emerge into.\n10. **Be patient and don't force it**: Respect your cat's boundaries and allow them to come out when they're ready. Forcing them to leave their hiding spot can create negative associations and make them more likely to hide in the future.\n\nRemember, every cat is different, so try a combination of these methods to see what works best for your feline friend." + } +] \ No newline at end of file diff --git a/Codes/ch07/05_dataset-generation/llama3-ollama.ipynb b/Codes/ch07/05_dataset-generation/llama3-ollama.ipynb new file mode 100644 index 0000000..812edf2 --- /dev/null +++ b/Codes/ch07/05_dataset-generation/llama3-ollama.ipynb @@ -0,0 +1,506 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "136a4efe-fb99-4311-8679-e0a5b6282755", + "metadata": {}, + "source": [ + "\n", + "\n", + "\n", + "\n", + "\n", + "
\n", + "\n", + "Supplementary code for the Build a Large Language Model From Scratch book by Sebastian Raschka
\n", + "
Code repository: https://github.com/rasbt/LLMs-from-scratch\n", + "\n", + "		\n", + "\n", + "
" + ] + }, + { + "cell_type": "markdown", + "id": "b1910a06-e8a3-40ac-8201-ff70615b1ba4", + "metadata": { + "tags": [] + }, + "source": [ + "# Generating An Instruction Dataset via Llama 3 and Ollama" + ] + }, + { + "cell_type": "markdown", + "id": "a128651b-f326-4232-a994-42f38b7ed520", + "metadata": {}, + "source": [ + "- This notebook uses an 8-billion-parameter Llama 3 model through ollama to generate a synthetic dataset using the \"hack\" proposed in the \"Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing\" paper ([https://arxiv.org/abs/2406.08464](https://arxiv.org/abs/2406.08464))\n", + "\n", + "- The generated dataset will be an instruction dataset with \"instruction\" and \"output\" field similar to what can be found in Alpaca:\n", + "\n", + "\n", + "```python\n", + "{\n", + " \"instruction\": \"What is the atomic number of helium?\",\n", + " \"output\": \"The atomic number of helium is 2.\",\n", + "},\n", + "```\n", + "\n", + "- The code doesn't require a GPU and runs on a laptop (it was tested on a M3 MacBook Air)\n", + "\n", + "*Note that the instruction datasets created here are for educational purposes. However, it is the users' duty to ensure that their use adheres to the terms of the relevant licensing agreements with Meta AI's Llama 3.*" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "63610acc-db94-437f-8d38-e99dca0299cb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "tqdm version: 4.66.4\n" + ] + } + ], + "source": [ + "from importlib.metadata import version\n", + "\n", + "pkgs = [\n", + " \"tqdm\", # Progress bar\n", + "]\n", + "\n", + "for p in pkgs:\n", + " print(f\"{p} version: {version(p)}\")" + ] + }, + { + "cell_type": "markdown", + "id": "8bcdcb34-ac75-4f4f-9505-3ce0666c42d5", + "metadata": {}, + "source": [ + "## Installing Ollama and Downloading Llama 3" + ] + }, + { + "cell_type": "markdown", + "id": "5a092280-5462-4709-a3fe-8669a4a8a0a6", + "metadata": {}, + "source": [ + "- Ollama is an application to run LLMs efficiently\n", + "- It is a wrapper around [llama.cpp](https://github.com/ggerganov/llama.cpp), which implements LLMs in pure C/C++ to maximize efficiency\n", + "- Note that it is a tool for using LLMs to generate text (inference), not training or finetuning LLMs\n", + "- Prior to running the code below, install ollama by visiting [https://ollama.com](https://ollama.com) and following the instructions (for instance, clicking on the \"Download\" button and downloading the ollama application for your operating system)" + ] + }, + { + "cell_type": "markdown", + "id": "9558a522-650d-401a-84fc-9fd7b1f39da7", + "metadata": {}, + "source": [ + "- For macOS and Windows users, click on the ollama application you downloaded; if it prompts you to install the command line usage, say \"yes\"\n", + "- Linux users can use the installation command provided on the ollama website\n", + "\n", + "- In general, before we can use ollama from the command line, we have to either start the ollama application or run `ollama serve` in a separate terminal\n", + "\n", + "\n", + "\n", + "\n", + "- With the ollama application or `ollama serve` running, in a different terminal, on the command line, execute the following command to try out the 8-billion-parameter Llama 3 model (the model, which takes up 4.7 GB of storage space, will be automatically downloaded the first time you execute this command)\n", + "\n", + "```bash\n", + "# 8B model\n", + "ollama run llama3\n", + "```\n", + "\n", + "\n", + "The output looks like as follows:\n", + "\n", + "```\n", + "$ ollama run llama3\n", + "pulling manifest \n", + "pulling 6a0746a1ec1a... 100% ▕████████████████▏ 4.7 GB                         \n", + "pulling 4fa551d4f938... 100% ▕████████████████▏  12 KB                         \n", + "pulling 8ab4849b038c... 100% ▕████████████████▏  254 B                         \n", + "pulling 577073ffcc6c... 100% ▕████████████████▏  110 B                         \n", + "pulling 3f8eb4da87fa... 100% ▕████████████████▏  485 B                         \n", + "verifying sha256 digest \n", + "writing manifest \n", + "removing any unused layers \n", + "success \n", + "```\n", + "\n", + "- Note that `llama3` refers to the instruction finetuned 8-billion-parameter Llama 3 model\n", + "\n", + "- Alternatively, you can also use the larger 70-billion-parameter Llama 3 model, if your machine supports it, by replacing `llama3` with `llama3:70b`\n", + "\n", + "- After the download has been completed, you will see a command line prompt that allows you to chat with the model\n", + "\n", + "- Try a prompt like \"What do llamas eat?\", which should return an output similar to the following:\n", + "\n", + "```\n", + ">>> What do llamas eat?\n", + "Llamas are ruminant animals, which means they have a four-chambered \n", + "stomach and eat plants that are high in fiber. In the wild, llamas \n", + "typically feed on:\n", + "1. Grasses: They love to graze on various types of grasses, including tall \n", + "grasses, wheat, oats, and barley.\n", + "```" + ] + }, + { + "cell_type": "markdown", + "id": "0b5addcb-fc7d-455d-bee9-6cc7a0d684c7", + "metadata": {}, + "source": [ + "- You can end this session using the input `/bye`" + ] + }, + { + "cell_type": "markdown", + "id": "dda155ee-cf36-44d3-b634-20ba8e1ca38a", + "metadata": {}, + "source": [ + "## Using Ollama's REST API" + ] + }, + { + "cell_type": "markdown", + "id": "89343a84-0ddc-42fc-bf50-298a342b93c0", + "metadata": {}, + "source": [ + "- Now, an alternative way to interact with the model is via its REST API in Python via the following function\n", + "- Before you run the next cells in this notebook, make sure that ollama is still running, as described above, via\n", + " - `ollama serve` in a terminal\n", + " - the ollama application\n", + "- Next, run the following code cell to query the model" + ] + }, + { + "cell_type": "markdown", + "id": "16642a48-1cab-40d2-af08-ab8c2fbf5876", + "metadata": {}, + "source": [ + "- First, let's try the API with a simple example to make sure it works as intended:" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "65b0ba76-1fb1-4306-a7c2-8f3bb637ccdb", + "metadata": {}, + "outputs": [], + "source": [ + "import urllib.request\n", + "import json\n", + "\n", + "def query_model(prompt, model=\"llama3\", url=\"http://localhost:11434/api/chat\", role=\"user\"):\n", + " # Create the data payload as a dictionary\n", + " data = {\n", + " \"model\": model,\n", + " \"seed\": 123, # for deterministic responses\n", + " \"temperature\": 1., # for deterministic responses\n", + " \"top_p\": 1, \n", + " \"messages\": [\n", + " {\"role\": role, \"content\": prompt}\n", + " ]\n", + " }\n", + "\n", + " # Convert the dictionary to a JSON formatted string and encode it to bytes\n", + " payload = json.dumps(data).encode(\"utf-8\")\n", + "\n", + " # Create a request object, setting the method to POST and adding necessary headers\n", + " request = urllib.request.Request(url, data=payload, method=\"POST\")\n", + " request.add_header(\"Content-Type\", \"application/json\")\n", + "\n", + " # Send the request and capture the response\n", + " response_data = \"\"\n", + " with urllib.request.urlopen(request) as response:\n", + " # Read and decode the response\n", + " while True:\n", + " line = response.readline().decode(\"utf-8\")\n", + " if not line:\n", + " break\n", + " response_json = json.loads(line)\n", + " response_data += response_json[\"message\"][\"content\"]\n", + "\n", + " return response_data" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "4fb61a4e-2706-431a-835e-7e472b42989e", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Llamas are herbivores, which means they primarily eat plants and plant-based foods. Their diet typically consists of:\n", + "\n", + "1. Grasses: Llamas love to graze on various types of grasses, including tall grasses, short grasses, and even weeds.\n", + "2. Hay: They enjoy eating hay, such as alfalfa or timothy hay, which provides them with fiber, protein, and other essential nutrients.\n", + "3. Grains: Llamas may eat grains like oats, barley, or corn as a supplement to their diet.\n", + "4. Leaves: They will also munch on leaves from trees and shrubs, including clover, alfalfa, and various types of leaves.\n", + "5. Fruits and vegetables: In the wild, llamas might eat fruits and vegetables that grow in their natural habitat, such as apples, carrots, or potatoes.\n", + "\n", + "In general, a llama's diet should consist of:\n", + "\n", + "* 50% grasses and hay\n", + "* 20% grains (like oats or corn)\n", + "* 10% leaves and other plant material\n", + "* 5% fruits and vegetables (as treats)\n", + "\n", + "It's essential to provide llamas with a balanced diet that meets their nutritional needs, as they can be prone to health issues if they don't receive the right combination of nutrients.\n" + ] + } + ], + "source": [ + "result = query_model(\"What do Llamas eat?\")\n", + "print(result)" + ] + }, + { + "cell_type": "markdown", + "id": "8c079c6c-5845-4b31-a648-060d0273cd1d", + "metadata": {}, + "source": [ + "## Extract Instructions" + ] + }, + { + "cell_type": "markdown", + "id": "92b09132-4a92-4492-9b67-24a644767498", + "metadata": {}, + "source": [ + "- Now, let's use the \"hack\" proposed in the paper: we provide the empty prompt template `\"<|begin_of_text|><|start_header_id|>user<|end_header_id|>\"` prompt, which will cause the instruction-finetuned Llama 3 model to generate an instruction" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "d7320a41-ed86-49e9-8eb1-5d609a82ad74", + "metadata": {}, + "outputs": [], + "source": [ + "def extract_instruction(text):\n", + " for content in text.split(\"\\n\"):\n", + " if content:\n", + " return content.strip()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "bc41b72f-a8cf-4367-b0ca-0bf8c1f094fd", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "I am trying to find a way to make my child's birthday party more special and unique. What are some creative ideas you have?\n" + ] + } + ], + "source": [ + "query = \"<|begin_of_text|><|start_header_id|>user<|end_header_id|>\"\n", + "\n", + "result = query_model(query, role=\"assistant\")\n", + "instruction = extract_instruction(result)\n", + "print(instruction)" + ] + }, + { + "cell_type": "markdown", + "id": "04d04ba7-bffc-47f0-87dc-d60fc676b14a", + "metadata": {}, + "source": [ + "- As we can see above, surprisingly, the model indeed generated an instruction" + ] + }, + { + "cell_type": "markdown", + "id": "162a4739-6f03-4092-a5c2-f57a0b6a4c4d", + "metadata": {}, + "source": [ + "## Generate Responses" + ] + }, + { + "cell_type": "markdown", + "id": "2542f8d3-2db2-4a89-ae50-8825eb19d3b6", + "metadata": {}, + "source": [ + "- Now, the next step is to create the corresponding response, which can be done by simply passing the instruction as input" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "2349eb06-710f-4459-8a03-1a3b2e1e8905", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "What an exciting question! I'd be delighted to help you come up with some creative and unique ideas to make your child's birthday party truly special!\n", + "\n", + "Here are a few ideas to get you started:\n", + "\n", + "1. **Themed Scavenger Hunt**: Plan a scavenger hunt based on the birthday child's favorite theme (e.g., superheroes, animals, or princesses). Hide clues and challenges throughout the party area, leading up to a final surprise.\n", + "2. **DIY Crafts Station**: Set up a craft station where kids can create their own party favors, such as customized t-shirts, crowns, or jewelry. This activity encourages creativity and makes for a memorable keepsake.\n", + "3. **Mystery Box Challenge**: Fill mystery boxes with different textures, smells, and sounds. Have the kids guess what's inside each box without looking. This game promotes problem-solving and teamwork.\n", + "4. **Indoor Camping Adventure**: Set up a cozy indoor \"camping\" area with sleeping bags, flashlights, and s'mores-making stations. Kids can enjoy a camping experience without leaving the party location.\n", + "5. **Personalized Photo Booth**: Create a customized photo booth with props and backdrops that match the birthday child's theme. This activity allows kids to take home special memories and share them on social media.\n", + "6. **Foodie Fun**: Plan a cooking or baking station where kids can make their own treats, such as cupcakes, pizzas, or trail mix. This activity teaches valuable skills and lets kids enjoy their creations.\n", + "7. **Outdoor Movie Night**: Set up an outdoor movie screen (or projector) with cozy seating and snacks. Screen the birthday child's favorite film or a classic kid-friendly movie.\n", + "8. **Science Experiments**: Host a science-themed party where kids can conduct fun experiments, such as making slime, creating lava lamps, or growing crystals.\n", + "9. **Karaoke Contest**: Set up a karaoke machine with popular kids' songs and have a singing competition. Offer prizes for the best performances, and provide fun props like microphones and costumes.\n", + "10. **Time Capsule Ceremony**: Have each guest bring a small item that represents their favorite memory or something they're looking forward to in the future. Bury the time capsule together as a group, with instructions to open it on a specific date (e.g., next year's birthday party).\n", + "11. **Special Guest Appearance**: Arrange for a special guest, such as a superhero, princess, or even a real-life animal (if feasible), to make an appearance at the party.\n", + "12. **Customized Games**: Design custom games and activities that fit the birthday child's interests and personality. This could include a customized version of a favorite game or a new game altogether.\n", + "\n", + "Remember, the key to making your child's birthday party unique is to incorporate elements that reflect their personality and interests. Mix and match these ideas or come up with something entirely new – the possibilities are endless!\n", + "\n", + "What do you think? Is there anything in particular that resonates with you, or would you like more suggestions?\n" + ] + } + ], + "source": [ + "response = query_model(instruction, role=\"user\")\n", + "print(response)" + ] + }, + { + "cell_type": "markdown", + "id": "b12cf92c-3272-4b36-ae30-d1135af56328", + "metadata": {}, + "source": [ + "## Generate Dataset" + ] + }, + { + "cell_type": "markdown", + "id": "470037f3-64f4-4465-9f00-55b69e883a04", + "metadata": {}, + "source": [ + "- We can scale up this approach to an arbitrary number of data samples (you may want to apply some optional filtering length or quality (e.g., using another LLM to rate the generated data)\n", + "- Below, we generate 5 synthetic instruction-response pairs, which takes about 3 minutes on an M3 MacBook Air\n", + "- (To generate a dataset suitable for instruction finetuning, we want to increase this to at least 1k to 50k and perhaps run it on a GPU to generate the examples in a more timely fashion)\n", + "\n", + "**Tip**\n", + "\n", + "- You can generate even higher-quality responses by changing `model=\"llama3\"` to `model=\"llama3:70b\"`, however, this will require more computational resources" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "3b9e94ab-02ef-4372-91cd-60128159fd83", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "100%|█████████████████████████████████████████████| 5/5 [02:37<00:00, 31.41s/it]\n" + ] + } + ], + "source": [ + "from tqdm import tqdm\n", + "\n", + "dataset_size = 5\n", + "dataset = []\n", + "\n", + "for i in tqdm(range(dataset_size)):\n", + "\n", + " result = query_model(query, role=\"assistant\")\n", + " instruction = extract_instruction(result)\n", + " response = query_model(instruction, role=\"user\")\n", + " entry = {\n", + " \"instruction\": instruction,\n", + " \"output\": response\n", + " }\n", + " dataset.append(entry)" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "5fdbc194-c12a-4138-96d1-51bf66ca1574", + "metadata": {}, + "outputs": [], + "source": [ + "with open(\"instruction-data-llama3-7b.json\", \"w\") as file:\n", + " json.dump(dataset, file, indent=4)" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "b4027ead-bba4-49b7-9965-47532c3fdeee", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[\n", + " {\n", + " \"instruction\": \"What is the significance of the number 7 in various cultures and religions?\",\n", + " \"output\": \"The number 7 has been a significant and recurring theme across many cultures and religions, often imbuing it with special meaning and symbolism. Here are some examples:\\n\\n1. **Numerology**: In numerology, the number 7 is considered sacred and mystical, associated with spiritual awakening, introspection, and enlightenment.\\n2. **Judaism**: The Torah has seven days of creation, seven weeks in the wilderness, and seven years of rest (Sabbatical year). Seven is also a symbol of completion or perfection.\\n3. **Christianity**: In Christianity, there are seven deadly sins, seven virtues, and seven sacraments. Jesus was said to have spoken seven sermons, and the number 7 appears in various biblical accounts, such as the seven days of creation and the seven angels who appear before God.\\n4. **Islam**: In Islamic tradition, there are seven heavens, seven earths, and seven veils that separate the physical world from the divine realm. The Quran mentions seven verses (Ayats) that were revealed to Muhammad in a single revelation.\\n5. **Hinduism**: In Hindu mythology, there are seven chakras (energy centers) in the human body, seven colors of the rainbow, and seven planets (including the sun and moon). The number 7 is also associated with the seven aspects of the divine feminine (Saptamatrikas).\\n6. **Buddhism**: In Buddhist cosmology, there are seven levels of existence (dhatus) and seven stages of enlightenment (bodhisattva path). The seven days of meditation retreats in Tibetan Buddhism are also a nod to this number.\\n7. **Ancient Greek and Roman cultures**: The number 7 was considered perfect or complete, with seven being the number of planets known at the time (including Earth, Sun, Moon, Mars, Mercury, Jupiter, and Venus). Seven days were also the typical duration of a week in these cultures.\\n8. **Chinese culture**: In Chinese astrology, there are seven celestial stems and 12 earthly branches, creating a system of cyclical time reckoning. The number 7 is also associated with good luck and prosperity.\\n9. **African cultures**: In some African traditions, the number 7 is associated with spiritual powers, such as divination and healing. Seven is also considered a sacred number in Yoruba culture, representing the seven orishas (deities).\\n10. **Esoteric traditions**: Many esoteric traditions, including Hermeticism, Gnosticism, and Theosophy, have employed the number 7 to represent spiritual principles, such as the seven planes of existence, seven chakras, or seven sacred seals.\\n\\nThese examples illustrate the widespread significance of the number 7 across various cultures and religions. While the specific meanings may vary, the number 7 often represents completion, perfection, spiritual growth, and mystical connections.\"\n", + " },\n", + " {\n", + " \"instruction\": \"I've been wondering, what is the most important thing you can learn from a child?\",\n", + " \"output\": \"What a great question!\\n\\nChildren have so much to teach us, and I think one of the most important things we can learn from them is the power of simplicity, curiosity, and joy. Children are often unencumbered by the complexities and worries that come with adulthood, and they approach life with a fresh perspective.\\n\\nHere are some specific things children can teach us:\\n\\n1. **Simplicity**: Children are masters of simplifying things. They don't get bogged down in details or worry about what others think. They just enjoy the moment and find joy in simple things like playing outside, drawing, or reading.\\n2. **Curiosity**: Kids are naturally curious and love to explore their world. They ask questions, investigate, and learn from experience. This curiosity can be a powerful reminder to us adults to stay open-minded and seek new knowledge and experiences.\\n3. **Joy and playfulness**: Children have an amazing capacity for joy and playfulness. They find happiness in the simplest things, like playing with blocks or having a water balloon fight. We could all benefit from embracing our inner child and finding more joy in life's simple pleasures.\\n4. **Unconditional love**: Children are capable of loving unconditionally, without judgment or expectation. This can be a powerful reminder to us adults to practice self-love, self-acceptance, and kindness towards others.\\n5. **Creativity and imagination**: Kids are incredibly creative and imaginative, often thinking outside the box and coming up with innovative solutions. This creativity can inspire us adults to think differently, try new things, and approach problems from unique angles.\\n6. **Resilience**: Children are surprisingly resilient in the face of challenges and setbacks. They learn to adapt, cope, and move forward, teaching us valuable lessons about perseverance and bouncing back from adversity.\\n\\nIn summary, learning from children can help us regain a sense of simplicity, curiosity, joy, and playfulness, while also reminding us of the importance of unconditional love, creativity, and resilience.\"\n", + " },\n", + " {\n", + " \"instruction\": null,\n", + " \"output\": \"\"\n", + " },\n", + " {\n", + " \"instruction\": \"What is the best way to deal with a difficult person?\",\n", + " \"output\": \"Dealing with a difficult person can be challenging, but there are strategies that can help you navigate the situation effectively. Here are some tips:\\n\\n1. **Stay calm**: Take a deep breath and try not to take their behavior personally. A calm demeanor can help de-escalate tensions and prevent misunderstandings.\\n2. **Listen actively**: Sometimes, people act out because they feel unheard or misunderstood. Make an effort to listen carefully to what they're saying, and respond thoughtfully.\\n3. **Set boundaries**: Establish clear limits on what you are and aren't willing to engage in. Be firm but respectful when communicating your needs.\\n4. **Avoid taking the bait**: Don't let their provocations get under your skin. Stay focused on the issue at hand and avoid getting drawn into an argument or debate.\\n5. **Use \\\"I\\\" statements**: When expressing yourself, use \\\"I\\\" statements instead of \\\"you\\\" statements, which can come across as accusatory. This helps to reduce defensiveness and promotes a more constructive conversation.\\n6. **Practice empathy**: Try to understand where the other person is coming from, even if you don't agree with their perspective. Showing that you care about their feelings can help diffuse tension.\\n7. **Don't take it personally**: Remember that the difficult person's behavior is often a reflection of themselves, not you. Keep your self-worth and confidence intact.\\n8. **Seek common ground**: Look for areas of agreement or shared interests. This can help to build bridges and create a more positive atmosphere.\\n9. **Use humor (carefully)**: A well-timed, lighthearted joke or witty remark can help diffuse tension and lighten the mood. However, be cautious not to offend or make light of serious issues.\\n10. **Know when to disengage**: If the situation becomes too heated or toxic, it may be necessary to take a step back and re-engage at a later time when emotions have cooled down.\\n11. **Seek support**: Don't be afraid to ask for help from friends, family, or a professional if you're struggling to manage your interactions with a difficult person.\\n12. **Practice self-care**: Take care of yourself physically, emotionally, and mentally. Engage in activities that bring you joy and help you maintain your energy and resilience.\\n\\nRemember, dealing with a difficult person is not about winning an argument or changing their behavior; it's about maintaining your own emotional well-being and responding constructively to the situation.\"\n", + " },\n", + " {\n", + " \"instruction\": \"I'm looking for a way to get my cat's attention when they're hiding under the bed or in a closet.\",\n", + " \"output\": \"The classic \\\"where'd my cat go?\\\" conundrum! Don't worry, I've got some tips to help you coax your kitty out from their hiding spots:\\n\\n1. **Use their favorite treats**: Cats love food, and familiar treats can be a powerful lure. Try calling your cat's name and saying \\\"treat time\\\" in a playful tone. This might encourage them to emerge and investigate.\\n2. **Make some noise**: Cats have poor eyesight but excellent hearing. Gently knock on the bed frame or closet door with your knuckles, making a soft, rhythmic sound. This can help your cat pinpoint where you are and entice them to come out.\\n3. **Speak softly and calmly**: When speaking to your cat, use a gentle, soothing tone. Avoid loud or harsh voices, as these might scare them even further into hiding.\\n4. **Use verbal cues**: Establish a consistent verbal cue, like \\\"come on out\\\" or \\\"let's play,\\\" which can become associated with the idea of leaving their hiding spot.\\n5. **Create a \\\"safe zone\\\"**: If your cat is hiding due to fear or anxiety (e.g., from loud noises or other pets), try creating a safe, comfortable space for them to emerge into. This could be a cozy blanket or a familiar toy.\\n6. **Patiently wait it out**: Sometimes, cats just need time and space to feel secure enough to come out. Give your cat the opportunity to leave their hiding spot at their own pace.\\n7. **Use a flashlight (carefully)**: If your cat is hiding in a dark space, try using a flashlight to create a gentle beam of light. Be cautious not to shine it directly into their eyes, as this could startle them further.\\n8. **Offer a familiar object**: Place a familiar toy or blanket near the entrance to the hiding spot, which can help your cat feel more comfortable coming out.\\n9. **Make the space inviting**: If your cat is hiding under the bed, try moving any clutter or dust bunnies away from the area. Make the space underneath the bed a pleasant place for them to emerge into.\\n10. **Be patient and don't force it**: Respect your cat's boundaries and allow them to come out when they're ready. Forcing them to leave their hiding spot can create negative associations and make them more likely to hide in the future.\\n\\nRemember, every cat is different, so try a combination of these methods to see what works best for your feline friend.\"\n", + " }\n", + "]" + ] + } + ], + "source": [ + "!cat instruction-data-llama3-7b.json" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.6" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/Codes/ch07/README.md b/Codes/ch07/README.md new file mode 100644 index 0000000..ca001aa --- /dev/null +++ b/Codes/ch07/README.md @@ -0,0 +1,15 @@ +# Chapter 7: Finetuning to Follow Instructions + +## Main Chapter Code + +- [01_main-chapter-code](01_main-chapter-code) contains the main chapter code and exercise solutions + +## Bonus Materials + +- [02_dataset-utilities](02_dataset-utilities) contains utility code that can be used for preparing an instruction dataset. + +- [03_model-evaluation](03_model-evaluation) contains utility code for evaluating instruction responses using a local Llama 3 model and the GPT-4 API. + +- [04_preference-tuning-with-dpo](04_preference-tuning-with-dpo) implements code for preference finetuning with Direct Preference Optimization (DPO) + +- [05_dataset-generation](05_dataset-generation) contains code to generate synthetic datasets for instruction finetuning