richarderkhov/abideen_-_monarchcoder-7b-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "https://cdn-uploads.huggingface.co/production/uploads/64e380b2e12618b261fa6ba0/oJN8_xoMOq2RlIc799m-x.jpeg",
    "summary": "!image/jpeg MonarchCoder-7B is a slerp merge of the following models using LazyMergekit: * Syed-Hasan-8503/Tess-Coder-7B-Mistral-v1.0 * mlabonne/AlphaMonarch-7B The main aim behind creating this model is to create a model that performs well in reasoning, conversation, and coding. AlphaMonarch pperforms amazing on reasoning and conversation tasks. Merging AlphaMonarch with a coding model yielded MonarchCoder-7B which performs better on OpenLLM, Nous, and HumanEval benchmark. Although MonarchCoder-2x7B performs better than MonarchCoder-7B.",
    "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\nMonarchCoder-7B - GGUF\n- Model creator: https://huggingface.co/abideen/\n- Original model: https://huggingface.co/abideen/MonarchCoder-7B/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [MonarchCoder-7B.Q2_K.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q2_K.gguf) | Q2_K | 2.01GB |\n| [MonarchCoder-7B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.IQ3_XS.gguf) | IQ3_XS | 2.81GB |\n| [MonarchCoder-7B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.IQ3_S.gguf) | IQ3_S | 2.96GB |\n| [MonarchCoder-7B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q3_K_S.gguf) | Q3_K_S | 2.95GB |\n| [MonarchCoder-7B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.IQ3_M.gguf) | IQ3_M | 3.06GB |\n| [MonarchCoder-7B.Q3_K.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q3_K.gguf) | Q3_K | 2.59GB |\n| [MonarchCoder-7B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q3_K_M.gguf) | Q3_K_M | 3.28GB |\n| [MonarchCoder-7B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q3_K_L.gguf) | Q3_K_L | 3.56GB |\n| [MonarchCoder-7B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.IQ4_XS.gguf) | IQ4_XS | 3.67GB |\n| [MonarchCoder-7B.Q4_0.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q4_0.gguf) | Q4_0 | 3.83GB |\n| [MonarchCoder-7B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.IQ4_NL.gguf) | IQ4_NL | 3.87GB |\n| [MonarchCoder-7B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q4_K_S.gguf) | Q4_K_S | 3.86GB |\n| [MonarchCoder-7B.Q4_K.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q4_K.gguf) | Q4_K | 4.07GB |\n| [MonarchCoder-7B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q4_K_M.gguf) | Q4_K_M | 4.07GB |\n| [MonarchCoder-7B.Q4_1.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q4_1.gguf) | Q4_1 | 4.24GB |\n| [MonarchCoder-7B.Q5_0.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q5_0.gguf) | Q5_0 | 4.65GB |\n| [MonarchCoder-7B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q5_K_S.gguf) | Q5_K_S | 4.65GB |\n| [MonarchCoder-7B.Q5_K.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q5_K.gguf) | Q5_K | 4.78GB |\n| [MonarchCoder-7B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q5_K_M.gguf) | Q5_K_M | 4.78GB |\n| [MonarchCoder-7B.Q5_1.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q5_1.gguf) | Q5_1 | 5.07GB |\n| [MonarchCoder-7B.Q6_K.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q6_K.gguf) | Q6_K | 5.53GB |\n| [MonarchCoder-7B.Q8_0.gguf](https://huggingface.co/RichardErkhov/abideen_-_MonarchCoder-7B-gguf/blob/main/MonarchCoder-7B.Q8_0.gguf) | Q8_0 | 7.17GB |\n\n\n\n\nOriginal model description:\n---\nlicense: apache-2.0\ntags:\n- merge\n- mergekit\n- lazymergekit\n- Syed-Hasan-8503/Tess-Coder-7B-Mistral-v1.0\n- mlabonne/AlphaMonarch-7B\nbase_model:\n- Syed-Hasan-8503/Tess-Coder-7B-Mistral-v1.0\n- mlabonne/AlphaMonarch-7B\nmodel-index:\n- name: MonarchCoder-7B\n  results:\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: AI2 Reasoning Challenge (25-Shot)\n      type: ai2_arc\n      config: ARC-Challenge\n      split: test\n      args:\n        num_few_shot: 25\n    metrics:\n    - type: acc_norm\n      value: 68.52\n      name: normalized accuracy\n    source:\n      url: >-\n        https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=abideen/MonarchCoder-7B\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: HellaSwag (10-Shot)\n      type: hellaswag\n      split: validation\n      args:\n        num_few_shot: 10\n    metrics:\n    - type: acc_norm\n      value: 87.3\n      name: normalized accuracy\n    source:\n      url: >-\n        https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=abideen/MonarchCoder-7B\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: MMLU (5-Shot)\n      type: cais/mmlu\n      config: all\n      split: test\n      args:\n        num_few_shot: 5\n    metrics:\n    - type: acc\n      value: 64.65\n      name: accuracy\n    source:\n      url: >-\n        https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=abideen/MonarchCoder-7B\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: TruthfulQA (0-shot)\n      type: truthful_qa\n      config: multiple_choice\n      split: validation\n      args:\n        num_few_shot: 0\n    metrics:\n    - type: mc2\n      value: 61.21\n    source:\n      url: >-\n        https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=abideen/MonarchCoder-7B\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: Winogrande (5-shot)\n      type: winogrande\n      config: winogrande_xl\n      split: validation\n      args:\n        num_few_shot: 5\n    metrics:\n    - type: acc\n      value: 80.19\n      name: accuracy\n    source:\n      url: >-\n        https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=abideen/MonarchCoder-7B\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: GSM8k (5-shot)\n      type: gsm8k\n      config: main\n      split: test\n      args:\n        num_few_shot: 5\n    metrics:\n    - type: acc\n      value: 65.13\n      name: accuracy\n    source:\n      url: >-\n        https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=abideen/MonarchCoder-7B\n      name: Open LLM Leaderboard\nlanguage:\n- en\nlibrary_name: transformers\n---\n\n# MonarchCoder-7B\n\n\n![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/64e380b2e12618b261fa6ba0/oJN8_xoMOq2RlIc799m-x.jpeg)\n\nMonarchCoder-7B is a slerp merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):\n* [Syed-Hasan-8503/Tess-Coder-7B-Mistral-v1.0](https://huggingface.co/Syed-Hasan-8503/Tess-Coder-7B-Mistral-v1.0)\n* [mlabonne/AlphaMonarch-7B](https://huggingface.co/mlabonne/AlphaMonarch-7B)\n\nThe main aim behind creating this model is to create a model that performs well in reasoning, conversation, and coding. AlphaMonarch pperforms amazing on reasoning and conversation tasks. Merging AlphaMonarch with a coding model yielded MonarchCoder-7B which performs better on OpenLLM, Nous, and HumanEval benchmark. Although [MonarchCoder-2x7B](abideen/MonarchCoder-MoE-2x7B) performs better than MonarchCoder-7B.\n\n\n## 🏆 Evaluation results\n```\n|             Metric              |MonarchCoder-Moe-2x7B||MonarchCoder-7B||AlphaMonarch|\n|---------------------------------|---------------------|-----------------|------------|\n|Avg.                             |       74.23         |      71.17      |   75.99    |\n|HumanEval                        |       41.15         |      39.02      |   34.14    |\n|HumanEval+                       |       29.87         |      31.70      |   29.26    |\n|MBPP                             |       40.60         |       *         |     *      |\n|AI2 Reasoning Challenge (25-Shot)|       70.99         |      68.52      |   73.04    |\n|HellaSwag (10-Shot)              |       87.99         |      87.30      |   89.18    |\n|MMLU (5-Shot)                    |       65.11         |      64.65      |   64.40    |\n|TruthfulQA (0-shot)              |       71.25         |      61.21      |   77.91    |\n|Winogrande (5-shot)              |       80.66         |      80.19     .|   84.69    |\n|GSM8k (5-shot)           .       |       69.37         |      65.13      |   66.72    | \n```\n\n## 🧩 Configuration\n\n```yaml\nslices:\n  - sources:\n      - model: Syed-Hasan-8503/Tess-Coder-7B-Mistral-v1.0\n        layer_range: [0, 32]\n      - model: mlabonne/AlphaMonarch-7B\n        layer_range: [0, 32]\nmerge_method: slerp\nbase_model: mlabonne/AlphaMonarch-7B\nparameters:\n  t:\n    - filter: self_attn\n      value: [0, 0.5, 0.3, 0.7, 1]\n    - filter: mlp\n      value: [1, 0.5, 0.7, 0.3, 0]\n    - value: 0.5\ndtype: bfloat16\n```\n\n## 💻 Usage\n\n```python\n!pip install -qU transformers accelerate\n\nfrom transformers import AutoTokenizer\nimport transformers\nimport torch\n\nmodel = \"abideen/MonarchCoder-7B\"\nmessages = [{\"role\": \"user\", \"content\": \"What is a large language model?\"}]\n\ntokenizer = AutoTokenizer.from_pretrained(model)\nprompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)\npipeline = transformers.pipeline(\n    \"text-generation\",\n    model=model,\n    torch_dtype=torch.float16,\n    device_map=\"auto\",\n)\n\noutputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)\nprint(outputs[0][\"generated_text\"])\n```\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 92,
  "gated": false,
  "private": false,
  "last_modified": "2024-09-09T02:41:12.000Z",
  "created_at": "2024-09-08T21:26:09.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66de16719a962ae594d1f14e",
  "id": "RichardErkhov/abideen_-_MonarchCoder-7B-gguf",
  "modelId": "RichardErkhov/abideen_-_MonarchCoder-7B-gguf",
  "sha": "625a8e6bc0841864a6dfeae46dd1b417a0bcfa2d",
  "createdAt": "2024-09-08T21:26:09.000Z",
  "lastModified": "2024-09-09T02:41:12.000Z",
  "author": "RichardErkhov",
  "downloads": 92,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}