richarderkhov/idea-ccnl_-_wenzhong-gpt2-110m-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "Quantization made by Richard Erkhov.\n\n[Github](https://github.com/RichardErkhov)\n\n[Discord](https://discord.gg/pvy7H8DZMG)\n\n[Request more models](https://github.com/RichardErkhov/quant_request)\n\n\nWenzhong-GPT2-110M - GGUF\n- Model creator: https://huggingface.co/IDEA-CCNL/\n- Original model: https://huggingface.co/IDEA-CCNL/Wenzhong-GPT2-110M/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Wenzhong-GPT2-110M.Q2_K.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q2_K.gguf) | Q2_K | 0.08GB |\n| [Wenzhong-GPT2-110M.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.IQ3_XS.gguf) | IQ3_XS | 0.08GB |\n| [Wenzhong-GPT2-110M.IQ3_S.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.IQ3_S.gguf) | IQ3_S | 0.08GB |\n| [Wenzhong-GPT2-110M.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q3_K_S.gguf) | Q3_K_S | 0.08GB |\n| [Wenzhong-GPT2-110M.IQ3_M.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.IQ3_M.gguf) | IQ3_M | 0.09GB |\n| [Wenzhong-GPT2-110M.Q3_K.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q3_K.gguf) | Q3_K | 0.09GB |\n| [Wenzhong-GPT2-110M.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q3_K_M.gguf) | Q3_K_M | 0.09GB |\n| [Wenzhong-GPT2-110M.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q3_K_L.gguf) | Q3_K_L | 0.1GB |\n| [Wenzhong-GPT2-110M.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.IQ4_XS.gguf) | IQ4_XS | 0.1GB |\n| [Wenzhong-GPT2-110M.Q4_0.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q4_0.gguf) | Q4_0 | 0.1GB |\n| [Wenzhong-GPT2-110M.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.IQ4_NL.gguf) | IQ4_NL | 0.1GB |\n| [Wenzhong-GPT2-110M.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q4_K_S.gguf) | Q4_K_S | 0.1GB |\n| [Wenzhong-GPT2-110M.Q4_K.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q4_K.gguf) | Q4_K | 0.11GB |\n| [Wenzhong-GPT2-110M.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q4_K_M.gguf) | Q4_K_M | 0.11GB |\n| [Wenzhong-GPT2-110M.Q4_1.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q4_1.gguf) | Q4_1 | 0.11GB |\n| [Wenzhong-GPT2-110M.Q5_0.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q5_0.gguf) | Q5_0 | 0.11GB |\n| [Wenzhong-GPT2-110M.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q5_K_S.gguf) | Q5_K_S | 0.11GB |\n| [Wenzhong-GPT2-110M.Q5_K.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q5_K.gguf) | Q5_K | 0.12GB |\n| [Wenzhong-GPT2-110M.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q5_K_M.gguf) | Q5_K_M | 0.12GB |\n| [Wenzhong-GPT2-110M.Q5_1.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q5_1.gguf) | Q5_1 | 0.12GB |\n| [Wenzhong-GPT2-110M.Q6_K.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q6_K.gguf) | Q6_K | 0.13GB |\n| [Wenzhong-GPT2-110M.Q8_0.gguf](https://huggingface.co/RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf/blob/main/Wenzhong-GPT2-110M.Q8_0.gguf) | Q8_0 | 0.17GB |\n\n\n\n\nOriginal model description:\n---\nlanguage: \n  - zh\n\ninference: \n  parameters:\n    temperature: 0.7\n    top_p: 0.6\n    repetition_penalty: 1.1\n    max_new_tokens: 128\n    num_return_sequences: 3\n    do_sample: true\n\nlicense: apache-2.0\ntags:\n- generate\n- gpt2\n\nwidget:\n- 北京是中国的\n- 西湖的景色\n\n---\n\n# Wenzhong-GPT2-110M\n\n- Main Page:[Fengshenbang](https://fengshenbang-lm.com/)\n- Github: [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM)\n## 简介 Brief Introduction\n\n善于处理NLG任务，中文版的GPT2-Small。\n\nFocused on handling NLG tasks, Chinese GPT2-Small.\n\n## 模型分类 Model Taxonomy\n\n|  需求 Demand  | 任务 Task       | 系列 Series      | 模型 Model    | 参数 Parameter | 额外 Extra |\n|  :----:  | :----:  | :----:  | :----:  | :----:  | :----:  |\n| 通用 General  | 自然语言生成 NLG | 闻仲 Wenzhong | GPT2 |      110M      |     中文 Chinese     |\n\n## 模型信息 Model Information\n\n类似于Wenzhong2.0-GPT2-3.5B-chinese，我们实现了一个small版本的12层的Wenzhong-GPT2-110M，并且在悟道（300G版本）上面进行预训练。\n\nSimilar to Wenzhong2.0-GPT2-3.5B-chinese, we implement a small size Wenzhong-GPT2-110M with 12 layers, which is pre-trained on Wudao Corpus (300G version).\n\n## 使用 Usage\n\n### 加载模型 Loading Models\n\n```python \nfrom transformers import GPT2Tokenizer,GPT2LMHeadModel\nhf_model_path = 'IDEA-CCNL/Wenzhong-GPT2-110M'\ntokenizer = GPT2Tokenizer.from_pretrained(hf_model_path)\nmodel = GPT2LMHeadModel.from_pretrained(hf_model_path)\n```\n\n### 使用示例 Usage Examples\n\n```python\nquestion = \"北京是中国的\"\ninputs = tokenizer(question,return_tensors='pt')\ngeneration_output = model.generate(**inputs,\n                                return_dict_in_generate=True,\n                                output_scores=True,\n                                max_length=150,\n                                # max_new_tokens=80,\n                                do_sample=True,\n                                top_p = 0.6,\n                                # num_beams=5,\n                                eos_token_id=50256,\n                                pad_token_id=0,\n                                num_return_sequences = 5)\n\nfor idx,sentence in enumerate(generation_output.sequences):\n    print('next sentence %d:\\n'%idx,\n    tokenizer.decode(sentence).split('<|endoftext|>')[0])\n    print('*'*40)\n```\n\n## 引用 Citation\n\n如果您在您的工作中使用了我们的模型，可以引用我们的[论文](https://arxiv.org/abs/2209.02970)：\n\nIf you are using the resource for your work, please cite the our [paper](https://arxiv.org/abs/2209.02970):\n\n```text\n@article{fengshenbang,\n  author    = {Jiaxing Zhang and Ruyi Gan and Junjie Wang and Yuxiang Zhang and Lin Zhang and Ping Yang and Xinyu Gao and Ziwei Wu and Xiaoqun Dong and Junqing He and Jianheng Zhuo and Qi Yang and Yongfeng Huang and Xiayu Li and Yanghan Wu and Junyu Lu and Xinyu Zhu and Weifeng Chen and Ting Han and Kunhao Pan and Rui Wang and Hao Wang and Xiaojun Wu and Zhongshen Zeng and Chongpei Chen},\n  title     = {Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence},\n  journal   = {CoRR},\n  volume    = {abs/2209.02970},\n  year      = {2022}\n}\n```\n\n也可以引用我们的[网站](https://github.com/IDEA-CCNL/Fengshenbang-LM/):\n\nYou can also cite our [website](https://github.com/IDEA-CCNL/Fengshenbang-LM/):\n\n```text\n@misc{Fengshenbang-LM,\n  title={Fengshenbang-LM},\n  author={IDEA-CCNL},\n  year={2021},\n  howpublished={\\url{https://github.com/IDEA-CCNL/Fengshenbang-LM}},\n}\n```\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "arxiv:2209.02970",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 248,
  "gated": false,
  "private": false,
  "last_modified": "2024-06-05T05:16:37.000Z",
  "created_at": "2024-06-05T05:04:35.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "665ff1e3cec3591d6b688f6e",
  "id": "RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf",
  "modelId": "RichardErkhov/IDEA-CCNL_-_Wenzhong-GPT2-110M-gguf",
  "sha": "3a663dc2e944fa7a72328a5cbda5544f099dc482",
  "createdAt": "2024-06-05T05:04:35.000Z",
  "lastModified": "2024-06-05T05:16:37.000Z",
  "author": "RichardErkhov",
  "downloads": 248,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}