richarderkhov/core-3_-_kuno-royale-v3-7b-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "Another experimental combination of eren23's ongo-monarch-jaskier merges and Kunoichi-DPO-v2-7B. Untested. kuno-royale-v3-7b is a merge of the following models using LazyMergekit: * SanjiWatsuki/Kunoichi-DPO-v2-7B * eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v3",
    "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\nkuno-royale-v3-7b - GGUF\n- Model creator: https://huggingface.co/core-3/\n- Original model: https://huggingface.co/core-3/kuno-royale-v3-7b/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [kuno-royale-v3-7b.Q2_K.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q2_K.gguf) | Q2_K | 2.53GB |\n| [kuno-royale-v3-7b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.IQ3_XS.gguf) | IQ3_XS | 2.81GB |\n| [kuno-royale-v3-7b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.IQ3_S.gguf) | IQ3_S | 2.96GB |\n| [kuno-royale-v3-7b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q3_K_S.gguf) | Q3_K_S | 2.95GB |\n| [kuno-royale-v3-7b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.IQ3_M.gguf) | IQ3_M | 3.06GB |\n| [kuno-royale-v3-7b.Q3_K.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q3_K.gguf) | Q3_K | 3.28GB |\n| [kuno-royale-v3-7b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q3_K_M.gguf) | Q3_K_M | 3.28GB |\n| [kuno-royale-v3-7b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q3_K_L.gguf) | Q3_K_L | 3.56GB |\n| [kuno-royale-v3-7b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.IQ4_XS.gguf) | IQ4_XS | 3.67GB |\n| [kuno-royale-v3-7b.Q4_0.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q4_0.gguf) | Q4_0 | 3.83GB |\n| [kuno-royale-v3-7b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.IQ4_NL.gguf) | IQ4_NL | 3.87GB |\n| [kuno-royale-v3-7b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q4_K_S.gguf) | Q4_K_S | 3.86GB |\n| [kuno-royale-v3-7b.Q4_K.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q4_K.gguf) | Q4_K | 4.07GB |\n| [kuno-royale-v3-7b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q4_K_M.gguf) | Q4_K_M | 4.07GB |\n| [kuno-royale-v3-7b.Q4_1.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q4_1.gguf) | Q4_1 | 4.24GB |\n| [kuno-royale-v3-7b.Q5_0.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q5_0.gguf) | Q5_0 | 4.65GB |\n| [kuno-royale-v3-7b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q5_K_S.gguf) | Q5_K_S | 4.65GB |\n| [kuno-royale-v3-7b.Q5_K.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q5_K.gguf) | Q5_K | 4.78GB |\n| [kuno-royale-v3-7b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q5_K_M.gguf) | Q5_K_M | 4.78GB |\n| [kuno-royale-v3-7b.Q5_1.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q5_1.gguf) | Q5_1 | 5.07GB |\n| [kuno-royale-v3-7b.Q6_K.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q6_K.gguf) | Q6_K | 5.53GB |\n| [kuno-royale-v3-7b.Q8_0.gguf](https://huggingface.co/RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf/blob/main/kuno-royale-v3-7b.Q8_0.gguf) | Q8_0 | 7.17GB |\n\n\n\n\nOriginal model description:\n---\nlicense: cc-by-nc-2.0\ntags:\n- merge\n- mergekit\n- lazymergekit\n- SanjiWatsuki/Kunoichi-DPO-v2-7B\n- eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v3\nbase_model:\n- SanjiWatsuki/Kunoichi-DPO-v2-7B\n- eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v3\nmodel-index:\n- name: kuno-royale-v3-7b\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: 71.76\n      name: normalized accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=core-3/kuno-royale-v3-7b\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: 88.23\n      name: normalized accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=core-3/kuno-royale-v3-7b\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: 65.06\n      name: accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=core-3/kuno-royale-v3-7b\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: 71.13\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=core-3/kuno-royale-v3-7b\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: 82.32\n      name: accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=core-3/kuno-royale-v3-7b\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: 70.81\n      name: accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=core-3/kuno-royale-v3-7b\n      name: Open LLM Leaderboard\n---\n\n# kuno-royale-v3-7b\n\nAnother experimental combination of eren23's ongo-monarch-jaskier merges and Kunoichi-DPO-v2-7B. Untested.\n\nkuno-royale-v3-7b is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):\n* [SanjiWatsuki/Kunoichi-DPO-v2-7B](https://huggingface.co/SanjiWatsuki/Kunoichi-DPO-v2-7B)\n* [eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v3](https://huggingface.co/eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v3)\n\n## 🧩 Configuration\n\n```yaml\nslices:\n  - sources:\n      - model: SanjiWatsuki/Kunoichi-DPO-v2-7B\n        layer_range: [0, 32]\n      - model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v3\n        layer_range: [0, 32]\nmerge_method: slerp\nbase_model: SanjiWatsuki/Kunoichi-DPO-v2-7B\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 = \"core-3/kuno-royale-v3-7b\"\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# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)\nDetailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_core-3__kuno-royale-v3-7b)\n\n|             Metric              |Value|\n|---------------------------------|----:|\n|Avg.                             |74.88|\n|AI2 Reasoning Challenge (25-Shot)|71.76|\n|HellaSwag (10-Shot)              |88.23|\n|MMLU (5-Shot)                    |65.06|\n|TruthfulQA (0-shot)              |71.13|\n|Winogrande (5-shot)              |82.32|\n|GSM8k (5-shot)                   |70.81|\n\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 186,
  "gated": false,
  "private": false,
  "last_modified": "2024-09-11T18:32:54.000Z",
  "created_at": "2024-09-11T12:38:47.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66e18f57418a59a0f7ffbbe4",
  "id": "RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf",
  "modelId": "RichardErkhov/core-3_-_kuno-royale-v3-7b-gguf",
  "sha": "2d750930b8552ba71af05e97a5a7aa289e6b24d6",
  "createdAt": "2024-09-11T12:38:47.000Z",
  "lastModified": "2024-09-11T18:32:54.000Z",
  "author": "RichardErkhov",
  "downloads": 186,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}