richarderkhov/embeddedllm_-_mistral-7b-merge-02-v0-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "This is an experiment to compare merging 2 models using DARE TIES versus SLERP 🦙 We are mainly interested to compare against Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp The 2 models involved in the merge as follows: 1. teknium/OpenHermes-2.5-Mistral-7B 2. Intel/neural-chat-7b-v3-3 The yaml config file for the merge is: ``yaml models: # no parameters necessary for base model parameters: weight: 0.5 density: 0.5 parameters: weight: 0.5 density: 0.5 merge_method: dare_ties base_model: mistralai/Mistral-7B-v0.1 parameters: int8_mask: true dtype: bfloat16 `` # Open LLM Leaderboard Note that with more tuning DARE TIES might achieve better results. |            | DARE TIES | SLERP | |------------|-----------|-------| | Average    | 70.69     | 71.38 | | ARC        | 67.49     | 68.09 | | HellaSwag  | 85.78     | 86.2  | | MMLU       | 64.1      | 64.26 | | TruthfulQA | 60.52     | 62.78 | | Winogrande | 79.01     | 79.16 | | GSM8K      | 67.25     | 67.78 |",
    "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\nMistral-7B-Merge-02-v0 - GGUF\n- Model creator: https://huggingface.co/EmbeddedLLM/\n- Original model: https://huggingface.co/EmbeddedLLM/Mistral-7B-Merge-02-v0/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Mistral-7B-Merge-02-v0.Q2_K.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q2_K.gguf) | Q2_K | 2.53GB |\n| [Mistral-7B-Merge-02-v0.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.IQ3_XS.gguf) | IQ3_XS | 2.81GB |\n| [Mistral-7B-Merge-02-v0.IQ3_S.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.IQ3_S.gguf) | IQ3_S | 2.96GB |\n| [Mistral-7B-Merge-02-v0.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q3_K_S.gguf) | Q3_K_S | 2.95GB |\n| [Mistral-7B-Merge-02-v0.IQ3_M.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.IQ3_M.gguf) | IQ3_M | 3.06GB |\n| [Mistral-7B-Merge-02-v0.Q3_K.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q3_K.gguf) | Q3_K | 3.28GB |\n| [Mistral-7B-Merge-02-v0.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q3_K_M.gguf) | Q3_K_M | 3.28GB |\n| [Mistral-7B-Merge-02-v0.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q3_K_L.gguf) | Q3_K_L | 3.56GB |\n| [Mistral-7B-Merge-02-v0.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.IQ4_XS.gguf) | IQ4_XS | 3.67GB |\n| [Mistral-7B-Merge-02-v0.Q4_0.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q4_0.gguf) | Q4_0 | 3.83GB |\n| [Mistral-7B-Merge-02-v0.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.IQ4_NL.gguf) | IQ4_NL | 3.87GB |\n| [Mistral-7B-Merge-02-v0.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q4_K_S.gguf) | Q4_K_S | 3.86GB |\n| [Mistral-7B-Merge-02-v0.Q4_K.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q4_K.gguf) | Q4_K | 4.07GB |\n| [Mistral-7B-Merge-02-v0.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q4_K_M.gguf) | Q4_K_M | 4.07GB |\n| [Mistral-7B-Merge-02-v0.Q4_1.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q4_1.gguf) | Q4_1 | 4.24GB |\n| [Mistral-7B-Merge-02-v0.Q5_0.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q5_0.gguf) | Q5_0 | 4.65GB |\n| [Mistral-7B-Merge-02-v0.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q5_K_S.gguf) | Q5_K_S | 4.65GB |\n| [Mistral-7B-Merge-02-v0.Q5_K.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q5_K.gguf) | Q5_K | 4.78GB |\n| [Mistral-7B-Merge-02-v0.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q5_K_M.gguf) | Q5_K_M | 4.78GB |\n| [Mistral-7B-Merge-02-v0.Q5_1.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q5_1.gguf) | Q5_1 | 5.07GB |\n| [Mistral-7B-Merge-02-v0.Q6_K.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q6_K.gguf) | Q6_K | 5.53GB |\n| [Mistral-7B-Merge-02-v0.Q8_0.gguf](https://huggingface.co/RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf/blob/main/Mistral-7B-Merge-02-v0.Q8_0.gguf) | Q8_0 | 7.17GB |\n\n\n\n\nOriginal model description:\n---\nlicense: apache-2.0\nlanguage:\n- en\ntags:\n- merge\nbase_model:\n- teknium/OpenHermes-2.5-Mistral-7B\n- Intel/neural-chat-7b-v3-3\n---\n\n# Model Description\nThis is an experiment to compare merging 2 models using DARE TIES versus SLERP 🦙\n\nWe are mainly interested to compare against [Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp](https://huggingface.co/Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp)\n\nThe 2 models involved in the merge as follows:\n1. [teknium/OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B)\n2. [Intel/neural-chat-7b-v3-3](https://huggingface.co/Intel/neural-chat-7b-v3-3)\n\n- base model: [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)\n\nThe yaml config file for the merge is:\n\n```yaml\nmodels:\n  - model: mistralai/Mistral-7B-v0.1\n    # no parameters necessary for base model\n  - model: teknium/OpenHermes-2.5-Mistral-7B\n    parameters:\n      weight: 0.5\n      density: 0.5\n  - model: Intel/neural-chat-7b-v3-3\n    parameters:\n      weight: 0.5\n      density: 0.5\nmerge_method: dare_ties\nbase_model: mistralai/Mistral-7B-v0.1\nparameters:\n  int8_mask: true\ndtype: bfloat16\n```\n\n# Open LLM Leaderboard\n\nNote that with more tuning DARE TIES might achieve better results.\n\n|            | DARE TIES | SLERP |\n|------------|-----------|-------|\n| Average    | 70.69     | 71.38 |\n| ARC        | 67.49     | 68.09 |\n| HellaSwag  | 85.78     | 86.2  |\n| MMLU       | 64.1      | 64.26 |\n| TruthfulQA | 60.52     | 62.78 |\n| Winogrande | 79.01     | 79.16 |\n| GSM8K      | 67.25     | 67.78 |\n\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 135,
  "gated": false,
  "private": false,
  "last_modified": "2024-07-27T00:55:10.000Z",
  "created_at": "2024-07-26T21:07:36.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66a41018d73bea6c16390015",
  "id": "RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf",
  "modelId": "RichardErkhov/EmbeddedLLM_-_Mistral-7B-Merge-02-v0-gguf",
  "sha": "92f0090ccb14066753b88c5d6686d41ba1569c1a",
  "createdAt": "2024-07-26T21:07:36.000Z",
  "lastModified": "2024-07-27T00:55:10.000Z",
  "author": "RichardErkhov",
  "downloads": 135,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}