richarderkhov/huihui-ai_-_meta-llama-3.1-8b-instruct-abliterated-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "This is an uncensored version of Llama 3.1 8B Instruct created with abliteration (see this article to know more about it). Special thanks to @FailSpy for the original code and technique. Please follow him if you're interested in abliterated models.",
    "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\nMeta-Llama-3.1-8B-Instruct-abliterated - GGUF\n- Model creator: https://huggingface.co/huihui-ai/\n- Original model: https://huggingface.co/huihui-ai/Meta-Llama-3.1-8B-Instruct-abliterated/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q2_K.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q2_K.gguf) | Q2_K | 2.96GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.IQ3_XS.gguf) | IQ3_XS | 3.28GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.IQ3_S.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.IQ3_S.gguf) | IQ3_S | 3.43GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K_S.gguf) | Q3_K_S | 3.41GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.IQ3_M.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.IQ3_M.gguf) | IQ3_M | 1.51GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K.gguf) | Q3_K | 3.74GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K_M.gguf) | Q3_K_M | 3.74GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q3_K_L.gguf) | Q3_K_L | 4.03GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.IQ4_XS.gguf) | IQ4_XS | 4.18GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q4_0.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q4_0.gguf) | Q4_0 | 4.34GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.IQ4_NL.gguf) | IQ4_NL | 4.38GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q4_K_S.gguf) | Q4_K_S | 4.37GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q4_K.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q4_K.gguf) | Q4_K | 4.58GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q4_K_M.gguf) | Q4_K_M | 4.58GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q4_1.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q4_1.gguf) | Q4_1 | 4.78GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q5_0.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q5_0.gguf) | Q5_0 | 5.21GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q5_K_S.gguf) | Q5_K_S | 5.21GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q5_K.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q5_K.gguf) | Q5_K | 5.34GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q5_K_M.gguf) | Q5_K_M | 5.34GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q5_1.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q5_1.gguf) | Q5_1 | 5.65GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q6_K.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q6_K.gguf) | Q6_K | 6.14GB |\n| [Meta-Llama-3.1-8B-Instruct-abliterated.Q8_0.gguf](https://huggingface.co/RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf/blob/main/Meta-Llama-3.1-8B-Instruct-abliterated.Q8_0.gguf) | Q8_0 | 7.95GB |\n\n\n\n\nOriginal model description:\n---\nlibrary_name: transformers\nlicense: llama3.1\nbase_model: meta-llama/Meta-Llama-3.1-8B-Instruct\ntags:\n- abliterated\n- uncensored\n---\n\n# 🦙 Meta-Llama-3.1-8B-Instruct-abliterated\n\n\nThis is an uncensored version of Llama 3.1 8B Instruct created with abliteration (see [this article](https://huggingface.co/blog/mlabonne/abliteration) to know more about it).\n\nSpecial thanks to [@FailSpy](https://huggingface.co/failspy) for the original code and technique. Please follow him if you're interested in abliterated models.\n\n## Evaluations\nThe following data has been re-evaluated and calculated as the average for each test.\n\n| Benchmark   | Llama-3.1-8b-Instruct | Meta-Llama-3.1-8B-Instruct-abliterated | \n|-------------|-----------------------|----------------------------------------| \n| IF_Eval     | **80.0**              | 78.98                                  | \n| MMLU Pro    | **36.34**             | 35.91                                  | \n| TruthfulQA  | 52.98                 | **55.42**                              | \n| BBH         | **48.72**             | 47.0                                   | \n| GPQA        | 33.55                 | **33.93**                              | \n\nThe script used for evaluation can be found inside this repository under /eval.sh, or click [here](https://huggingface.co/huihui-ai/Meta-Llama-3.1-8B-Instruct-abliterated/blob/main/eval.sh)\n\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 183,
  "gated": false,
  "private": false,
  "last_modified": "2024-10-05T02:56:26.000Z",
  "created_at": "2024-10-04T20:45:45.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "670053f96537491c73c9c7b1",
  "id": "RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf",
  "modelId": "RichardErkhov/huihui-ai_-_Meta-Llama-3.1-8B-Instruct-abliterated-gguf",
  "sha": "fffb00ad5cdbcc6c458e0852fa2e59c088a75e15",
  "createdAt": "2024-10-04T20:45:45.000Z",
  "lastModified": "2024-10-05T02:56:26.000Z",
  "author": "RichardErkhov",
  "downloads": 183,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}