richarderkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "The following were merged with DARE using https://github.com/martyn/safetensors-merge-supermario",
    "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\nllama2-megamerge-dare-13b-v2 - GGUF\n- Model creator: https://huggingface.co/martyn/\n- Original model: https://huggingface.co/martyn/llama2-megamerge-dare-13b-v2/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [llama2-megamerge-dare-13b-v2.Q2_K.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q2_K.gguf) | Q2_K | 4.52GB |\n| [llama2-megamerge-dare-13b-v2.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.IQ3_XS.gguf) | IQ3_XS | 4.99GB |\n| [llama2-megamerge-dare-13b-v2.IQ3_S.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.IQ3_S.gguf) | IQ3_S | 5.27GB |\n| [llama2-megamerge-dare-13b-v2.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q3_K_S.gguf) | Q3_K_S | 5.27GB |\n| [llama2-megamerge-dare-13b-v2.IQ3_M.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.IQ3_M.gguf) | IQ3_M | 5.57GB |\n| [llama2-megamerge-dare-13b-v2.Q3_K.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q3_K.gguf) | Q3_K | 5.9GB |\n| [llama2-megamerge-dare-13b-v2.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q3_K_M.gguf) | Q3_K_M | 5.9GB |\n| [llama2-megamerge-dare-13b-v2.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q3_K_L.gguf) | Q3_K_L | 6.45GB |\n| [llama2-megamerge-dare-13b-v2.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.IQ4_XS.gguf) | IQ4_XS | 6.54GB |\n| [llama2-megamerge-dare-13b-v2.Q4_0.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q4_0.gguf) | Q4_0 | 6.86GB |\n| [llama2-megamerge-dare-13b-v2.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.IQ4_NL.gguf) | IQ4_NL | 6.9GB |\n| [llama2-megamerge-dare-13b-v2.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q4_K_S.gguf) | Q4_K_S | 6.91GB |\n| [llama2-megamerge-dare-13b-v2.Q4_K.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q4_K.gguf) | Q4_K | 7.33GB |\n| [llama2-megamerge-dare-13b-v2.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q4_K_M.gguf) | Q4_K_M | 7.33GB |\n| [llama2-megamerge-dare-13b-v2.Q4_1.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q4_1.gguf) | Q4_1 | 7.61GB |\n| [llama2-megamerge-dare-13b-v2.Q5_0.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q5_0.gguf) | Q5_0 | 8.36GB |\n| [llama2-megamerge-dare-13b-v2.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q5_K_S.gguf) | Q5_K_S | 8.36GB |\n| [llama2-megamerge-dare-13b-v2.Q5_K.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q5_K.gguf) | Q5_K | 8.6GB |\n| [llama2-megamerge-dare-13b-v2.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q5_K_M.gguf) | Q5_K_M | 8.6GB |\n| [llama2-megamerge-dare-13b-v2.Q5_1.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q5_1.gguf) | Q5_1 | 9.1GB |\n| [llama2-megamerge-dare-13b-v2.Q6_K.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q6_K.gguf) | Q6_K | 9.95GB |\n| [llama2-megamerge-dare-13b-v2.Q8_0.gguf](https://huggingface.co/RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf/blob/main/llama2-megamerge-dare-13b-v2.Q8_0.gguf) | Q8_0 | 12.88GB |\n\n\n\n\nOriginal model description:\n---\nlicense: llama2\nlanguage:\n- en\npipeline_tag: text-generation\ninference: false\ntags:\n- dare\n- super mario merge\n- pytorch\n- llama\n- llama-2\n- merge\n---\n\n# llama 13b mega merge v2\n\nThe following were merged with DARE using [https://github.com/martyn/safetensors-merge-supermario](https://github.com/martyn/safetensors-merge-supermario)\n\n## Mergelist\n\n```\nmeta-llama/Llama-2-13b-hf\najibawa-2023/Code-13B\najibawa-2023/Python-Code-13B\nmeta-math/MetaMath-13B-V1.0\nrombodawg/LosslessMegaCoder-llama2-13b-mini\nNousResearch/Nous-Hermes-Llama2-13b\nallenai/digital-socrates-13b\nmigtissera/Synthia-13B\nGryphe/MythoLogic-L2-13b\nallenai/tulu-2-dpo-13b\nFPHam/Free_Sydney_13b_HF\nFPHam/Free_Sydney_V2_13b_HF\nFPHam/Sydney_Overthinker_13b_HF\nKoboldAI/LLaMA2-13B-Psyfighter2\nUndi95/Unholy-v1-12L-13B\nathirdpath/Eileithyia-13B\nathirdpath/Orca-2-13b-Alpaca-Uncensored\n```\n\n## Merge command\n\n```\npython hf_merge.py mergelist.txt 13b-merge-v2 -p 0.11 -lambda 2.1\n```\n\n## Notes\n\n* seems to generalize instruct styles\n* `p` and `lambda` are still a guess\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 160,
  "gated": false,
  "private": false,
  "last_modified": "2024-08-03T17:45:35.000Z",
  "created_at": "2024-08-03T12:44:48.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66ae2640c7a575aa0e1f3815",
  "id": "RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf",
  "modelId": "RichardErkhov/martyn_-_llama2-megamerge-dare-13b-v2-gguf",
  "sha": "0034c6a5aa14c4f12373cbca13544bc5e85018a0",
  "createdAt": "2024-08-03T12:44:48.000Z",
  "lastModified": "2024-08-03T17:45:35.000Z",
  "author": "RichardErkhov",
  "downloads": 160,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}