richarderkhov/anthracite-org_-_magnum-v2-12b-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png",
    "summary": "Detailed results can be found here |      Metric       |Value| |-------------------|----:| |Avg.               |18.68| |IFEval (0-Shot)    |37.62| |BBH (3-Shot)       |28.79| |MATH Lvl 5 (4-Shot)| 4.76| |GPQA (0-shot)      | 5.48| |MuSR (0-shot)      |11.37| |MMLU-PRO (5-shot)  |24.08|",
    "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\nmagnum-v2-12b - GGUF\n- Model creator: https://huggingface.co/anthracite-org/\n- Original model: https://huggingface.co/anthracite-org/magnum-v2-12b/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [magnum-v2-12b.Q2_K.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q2_K.gguf) | Q2_K | 4.46GB |\n| [magnum-v2-12b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.IQ3_XS.gguf) | IQ3_XS | 4.94GB |\n| [magnum-v2-12b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.IQ3_S.gguf) | IQ3_S | 5.18GB |\n| [magnum-v2-12b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q3_K_S.gguf) | Q3_K_S | 5.15GB |\n| [magnum-v2-12b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.IQ3_M.gguf) | IQ3_M | 5.33GB |\n| [magnum-v2-12b.Q3_K.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q3_K.gguf) | Q3_K | 5.67GB |\n| [magnum-v2-12b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q3_K_M.gguf) | Q3_K_M | 5.67GB |\n| [magnum-v2-12b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q3_K_L.gguf) | Q3_K_L | 6.11GB |\n| [magnum-v2-12b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.IQ4_XS.gguf) | IQ4_XS | 6.33GB |\n| [magnum-v2-12b.Q4_0.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q4_0.gguf) | Q4_0 | 6.59GB |\n| [magnum-v2-12b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.IQ4_NL.gguf) | IQ4_NL | 6.65GB |\n| [magnum-v2-12b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q4_K_S.gguf) | Q4_K_S | 6.63GB |\n| [magnum-v2-12b.Q4_K.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q4_K.gguf) | Q4_K | 6.96GB |\n| [magnum-v2-12b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q4_K_M.gguf) | Q4_K_M | 6.96GB |\n| [magnum-v2-12b.Q4_1.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q4_1.gguf) | Q4_1 | 7.26GB |\n| [magnum-v2-12b.Q5_0.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q5_0.gguf) | Q5_0 | 7.93GB |\n| [magnum-v2-12b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q5_K_S.gguf) | Q5_K_S | 7.93GB |\n| [magnum-v2-12b.Q5_K.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q5_K.gguf) | Q5_K | 8.13GB |\n| [magnum-v2-12b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q5_K_M.gguf) | Q5_K_M | 8.13GB |\n| [magnum-v2-12b.Q5_1.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q5_1.gguf) | Q5_1 | 8.61GB |\n| [magnum-v2-12b.Q6_K.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q6_K.gguf) | Q6_K | 9.37GB |\n| [magnum-v2-12b.Q8_0.gguf](https://huggingface.co/RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf/blob/main/magnum-v2-12b.Q8_0.gguf) | Q8_0 | 12.13GB |\n\n\n\n\nOriginal model description:\n---\nlanguage:\n- en\n- fr\n- de\n- es\n- it\n- pt\n- ru\n- zh\n- ja\nlicense: apache-2.0\ntags:\n- chat\nbase_model: mistralai/Mistral-Nemo-Base-2407\npipeline_tag: text-generation\nmodel-index:\n- name: magnum-v2-12b\n  results:\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: IFEval (0-Shot)\n      type: HuggingFaceH4/ifeval\n      args:\n        num_few_shot: 0\n    metrics:\n    - type: inst_level_strict_acc and prompt_level_strict_acc\n      value: 37.62\n      name: strict accuracy\n    source:\n      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v2-12b\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: BBH (3-Shot)\n      type: BBH\n      args:\n        num_few_shot: 3\n    metrics:\n    - type: acc_norm\n      value: 28.79\n      name: normalized accuracy\n    source:\n      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v2-12b\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: MATH Lvl 5 (4-Shot)\n      type: hendrycks/competition_math\n      args:\n        num_few_shot: 4\n    metrics:\n    - type: exact_match\n      value: 4.76\n      name: exact match\n    source:\n      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v2-12b\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: GPQA (0-shot)\n      type: Idavidrein/gpqa\n      args:\n        num_few_shot: 0\n    metrics:\n    - type: acc_norm\n      value: 5.48\n      name: acc_norm\n    source:\n      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v2-12b\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: MuSR (0-shot)\n      type: TAUR-Lab/MuSR\n      args:\n        num_few_shot: 0\n    metrics:\n    - type: acc_norm\n      value: 11.37\n      name: acc_norm\n    source:\n      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v2-12b\n      name: Open LLM Leaderboard\n  - task:\n      type: text-generation\n      name: Text Generation\n    dataset:\n      name: MMLU-PRO (5-shot)\n      type: TIGER-Lab/MMLU-Pro\n      config: main\n      split: test\n      args:\n        num_few_shot: 5\n    metrics:\n    - type: acc\n      value: 24.08\n      name: accuracy\n    source:\n      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=anthracite-org/magnum-v2-12b\n      name: Open LLM Leaderboard\n---\n\n![image/png](https://cdn-uploads.huggingface.co/production/uploads/658a46cbfb9c2bdfae75b3a6/A9n8EJBDQziJWnXhOYeEE.png)\n\nThis is the fourth in a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus. This model is fine-tuned on top of [Mistral-Nemo-Base-2407](https://huggingface.co/mistralai/Mistral-Nemo-Base-2407).\n\n## Prompting\nModel has been Instruct tuned with the ChatML formatting. A typical input would look like this:\n\n```py\n\"\"\"<|im_start|>system\nsystem prompt<|im_end|>\n<|im_start|>user\nHi there!<|im_end|>\n<|im_start|>assistant\nNice to meet you!<|im_end|>\n<|im_start|>user\nCan I ask a question?<|im_end|>\n<|im_start|>assistant\n\"\"\"\n```\n\n## Credits\n- Stheno dataset (filtered)\n- [kalomaze/Opus_Instruct_25k](https://huggingface.co/datasets/kalomaze/Opus_Instruct_25k)\n- [Nopm/Opus_WritingStruct](https://huggingface.co/datasets/Nopm/Opus_WritingStruct)\n- [Gryphe/Sonnet3.5-SlimOrcaDedupCleaned](https://huggingface.co/datasets/Gryphe/Sonnet3.5-SlimOrcaDedupCleaned) (A ~16k rows subset)\n- [kalomaze/Opus_Instruct_3k](https://huggingface.co/datasets/kalomaze/Opus_Instruct_3k)\n\nThis model has been a team effort, and the credits goes to all members of Anthracite.\n\n## Training\nThe training was done for 2 epochs. We used 8x [NVIDIA H100 Tensor Core](https://www.nvidia.com/en-us/data-center/h100/) GPUs for the full-parameter fine-tuning of the model.\n\n[<img src=\"https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png\" alt=\"Built with Axolotl\" width=\"200\" height=\"32\"/>](https://github.com/OpenAccess-AI-Collective/axolotl)\n\n## Safety\n...\n# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)\nDetailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_anthracite-org__magnum-v2-12b)\n\n|      Metric       |Value|\n|-------------------|----:|\n|Avg.               |18.68|\n|IFEval (0-Shot)    |37.62|\n|BBH (3-Shot)       |28.79|\n|MATH Lvl 5 (4-Shot)| 4.76|\n|GPQA (0-shot)      | 5.48|\n|MuSR (0-shot)      |11.37|\n|MMLU-PRO (5-shot)  |24.08|\n\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 81,
  "gated": false,
  "private": false,
  "last_modified": "2024-09-17T19:27:40.000Z",
  "created_at": "2024-09-17T14:33:20.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66e993309c89a166b18569b4",
  "id": "RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf",
  "modelId": "RichardErkhov/anthracite-org_-_magnum-v2-12b-gguf",
  "sha": "9fc1bbd2db394a6b6be14186077304da292d8ec6",
  "createdAt": "2024-09-17T14:33:20.000Z",
  "lastModified": "2024-09-17T19:27:40.000Z",
  "author": "RichardErkhov",
  "downloads": 81,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}