llms-from-scratch-cn/Model_Architecture_Discussions/olmo/olmo.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0364fa99-3cad-4c11-ac41-6523fb98d187",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "env: HF_ENDPOINT=https://hf-mirror.com\n"
     ]
    }
   ],
   "source": [
    "%env HF_ENDPOINT=https://hf-mirror.com"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c654b825-84fd-43df-8412-53b1f9ecb8c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "# 设置 HF_HOME 环境变量 设置下载路径\n",
    "os.environ['HF_HOME'] = '/data1/ckw'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f30fc135-f12f-43bd-96e3-7ab02ef91296",
   "metadata": {},
   "outputs": [],
   "source": [
    "# %pip install jieba -q"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "94abdb98-fb74-42c0-805b-03df9fd12311",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "37fa141231cf48ec9c6e6e60c8c692cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model.safetensors:  16%|#5        | 744M/4.71G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "517e4c72255340fbb08e34a6bddbd0ce",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "generation_config.json:   0%|          | 0.00/116 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "35428e7696bf4df8aa1ffc047b1c8b39",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer_config.json:   0%|          | 0.00/493 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5a318d8c3cbc465382ec4b422909a925",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer.json: 0.00B [00:00, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "768ac1868dd1498a82c969c1abc688dc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "special_tokens_map.json:   0%|          | 0.00/65.0 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'\\nDataWhalechina is an organization founded at Shanghai Jiao Tong University that helps learners learn artificial intelligence.\\nThe company has developed a platform'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import AutoTokenizer\n",
    "from modeling_olmo import OlmoForCausalLM\n",
    "\n",
    "model = OlmoForCausalLM.from_pretrained(\"allenai/OLMo-1B-hf\")\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"allenai/OLMo-1B-hf\")\n",
    "\n",
    "prompt = '\\nDataWhalechina is an organization founded at Shanghai Jiao Tong University that helps learners learn artificial intelligence.'\n",
    "inputs = tokenizer(prompt, return_tensors=\"pt\")\n",
    "\n",
    "# Generate\n",
    "generate_ids = model.generate(inputs.input_ids, max_length=30)\n",
    "tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "61c6597d-3cdd-4a07-aa21-5b27e2cb6914",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nDataWhalechina is an organization founded at Shanghai Jiao Tong University that helps learners learn artificial intelligence.\\nThe company has developed a platform that allows users to learn AI by watching videos and answering questions.\\nThe platform is designed to be easy to use and has a variety of features that make it a great choice for learners.\\nDataWhalechina is a platform that helps learners learn artificial intelligence.\\nThe platform is designed to be easy to use and has a variety of features that make it a great choice for learners.\\nThe platform is also designed to be flexible, so that it can be used by different types of learners.\\nDataWhalechina is a platform that helps learners learn artificial intelligence.\\nThe platform is designed to be easy to use and has a variety of features that make it a great choice for learners.\\nThe platform is also designed to be flexible, so that it can be used by different types of learners.\\nDataWhalechina is a platform that helps learners learn artificial intelligence.\\nThe platform is designed to be easy to use and has a variety of features that make it a great choice for learners.\\nThe platform is also designed to be flexible, so that it can be used by different types of learners.\\nDataWhalechina is a platform that helps learners learn artificial intelligence.\\nThe platform is designed to be easy to use and has a variety of features that make it a great choice for learners.\\nThe'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "generate_ids = model.generate(inputs.input_ids, max_length=300)\n",
    "tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6c503178-b46b-445d-9555-bb529acecb47",
   "metadata": {},
   "outputs": [],
   "source": [
    "olmo而是一款不错的模型。生成速度也比较快"
   ]
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}