richarderkhov/evillain_-_stardust_20b_v0.2-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "GGUF Version: StarDust_20B_v0.2-GGUF Exllamav2 4.55bpw: Evillain/StarDust_20B_v0.2_exl2_4.55bpw Experimental merge of 3 capable models. Main purpose was to keep the DarkForest adventuring quality while make it more consistent with prompt and context following, and also increase RP and dialogues variety. This model is suitable for adventuring, should be fine also for storytelling, RP and ERP, but I don't have much of a time and imagination for testing. I haven't tested this model much, actually, so it's all about an exploration of it's capabilities. # Prompt Should work well with Alpaca format(but I'm not sure, since I don't really know what I'm doing :D) `` Below is an instruction that describes a task. Write a response that appropriately completes the request. ### Instruction: {prompt} ### Input: {input} ### Response: `` DaringMaid was tested with Undi/Ikaris SillyTavern presets for Noromaid: Context template, Instruct template, so maybe it have a sense to try my model with those templates too. # Merged This is a merge of pre-trained language models created using mergekit.",
    "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\nStarDust_20B_v0.2 - GGUF\n- Model creator: https://huggingface.co/Evillain/\n- Original model: https://huggingface.co/Evillain/StarDust_20B_v0.2/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [StarDust_20B_v0.2.Q2_K.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q2_K.gguf) | Q2_K | 6.91GB |\n| [StarDust_20B_v0.2.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.IQ3_XS.gguf) | IQ3_XS | 7.63GB |\n| [StarDust_20B_v0.2.IQ3_S.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.IQ3_S.gguf) | IQ3_S | 8.06GB |\n| [StarDust_20B_v0.2.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q3_K_S.gguf) | Q3_K_S | 8.06GB |\n| [StarDust_20B_v0.2.IQ3_M.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.IQ3_M.gguf) | IQ3_M | 8.53GB |\n| [StarDust_20B_v0.2.Q3_K.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q3_K.gguf) | Q3_K | 9.04GB |\n| [StarDust_20B_v0.2.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q3_K_M.gguf) | Q3_K_M | 9.04GB |\n| [StarDust_20B_v0.2.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q3_K_L.gguf) | Q3_K_L | 9.9GB |\n| [StarDust_20B_v0.2.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.IQ4_XS.gguf) | IQ4_XS | 10.01GB |\n| [StarDust_20B_v0.2.Q4_0.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q4_0.gguf) | Q4_0 | 10.52GB |\n| [StarDust_20B_v0.2.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.IQ4_NL.gguf) | IQ4_NL | 10.57GB |\n| [StarDust_20B_v0.2.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q4_K_S.gguf) | Q4_K_S | 7.56GB |\n| [StarDust_20B_v0.2.Q4_K.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q4_K.gguf) | Q4_K | 5.57GB |\n| [StarDust_20B_v0.2.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q4_K_M.gguf) | Q4_K_M | 4.12GB |\n| [StarDust_20B_v0.2.Q4_1.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q4_1.gguf) | Q4_1 | 3.66GB |\n| [StarDust_20B_v0.2.Q5_0.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q5_0.gguf) | Q5_0 | 2.97GB |\n| [StarDust_20B_v0.2.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q5_K_S.gguf) | Q5_K_S | 2.04GB |\n| [StarDust_20B_v0.2.Q5_K.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q5_K.gguf) | Q5_K | 1.65GB |\n| [StarDust_20B_v0.2.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q5_K_M.gguf) | Q5_K_M | 1.39GB |\n| [StarDust_20B_v0.2.Q5_1.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q5_1.gguf) | Q5_1 | 1.31GB |\n| [StarDust_20B_v0.2.Q6_K.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q6_K.gguf) | Q6_K | 1.05GB |\n| [StarDust_20B_v0.2.Q8_0.gguf](https://huggingface.co/RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf/blob/main/StarDust_20B_v0.2.Q8_0.gguf) | Q8_0 | 0.99GB |\n\n\n\n\nOriginal model description:\n---\nlicense: other\nlibrary_name: transformers\ntags:\n- mergekit\n- merge\n- not-for-all-audiences\nbase_model:\n- Kooten/DaringMaid-20B\n- TeeZee/DarkForest-20B-v2.0\n- athirdpath/Iambe-RP-v3-20b\nlicense_name: microsoft-research-license\nmodel-index:\n- name: StarDust_20B_v0.2\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: 61.01\n      name: normalized accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Evillain/StarDust_20B_v0.2\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: 83.76\n      name: normalized accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Evillain/StarDust_20B_v0.2\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: 59.29\n      name: accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Evillain/StarDust_20B_v0.2\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: 51.43\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Evillain/StarDust_20B_v0.2\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: 77.27\n      name: accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Evillain/StarDust_20B_v0.2\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: 24.03\n      name: accuracy\n    source:\n      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Evillain/StarDust_20B_v0.2\n      name: Open LLM Leaderboard\n---\n\n# Model Info\n\nGGUF Version: [StarDust_20B_v0.2-GGUF](https://huggingface.co/Evillain/StarDust_20B_v0.2-GGUF)\n \nExllamav2 4.55bpw: [Evillain/StarDust_20B_v0.2_exl2_4.55bpw](https://huggingface.co/Evillain/StarDust_20B_v0.2_exl2_4.55bpw)\n\nExperimental merge of 3 capable models. Main purpose was to keep the DarkForest adventuring quality while make it more consistent with prompt and context following, and also increase RP and dialogues variety.\n\nThis model is suitable for adventuring, should be fine also for storytelling, RP and ERP, but I don't have much of a time and imagination for testing.\n\nI haven't tested this model much, actually, so it's all about an exploration of it's capabilities.\n\n# Prompt\n\nShould work well with Alpaca format(but I'm not sure, since I don't really know what I'm doing :D)\n\n```\nBelow is an instruction that describes a task. Write a response that appropriately completes the request.\n### Instruction:\n{prompt}\n### Input:\n{input}\n### Response:\n```\n\nDaringMaid was tested with Undi/Ikaris SillyTavern presets for Noromaid: [Context template](https://files.catbox.moe/l5hguf.json), [Instruct template](https://files.catbox.moe/nzygwn.json), so maybe it have a sense to try my model with those templates too.\n\n# Merged\n\nThis is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).\n\n## Merge Details\n\nMy second ever merge, created in order to understand how everything works.\n\nFirstly I merged [athirdpath/Iambe-RP-v3-20b](https://huggingface.co/athirdpath/Iambe-RP-v3-20b) and [Kooten/DaringMaid-20B](https://huggingface.co/Kooten/DaringMaid-20B), named it Dust and repeated the operation with [TeeZee/DarkForest-20B-v2.0](https://huggingface.co/TeeZee/DarkForest-20B-v2.0) as a base and Dust as second model.\n\n### Merge Method\n\nThis model was merged using the SLERP merge method.\n\n### Models Merged\n\nThe following models were included in the merge:\n* [athirdpath/Iambe-RP-v3-20b](https://huggingface.co/athirdpath/Iambe-RP-v3-20b)\n* [Kooten/DaringMaid-20B](https://huggingface.co/Kooten/DaringMaid-20B)\n* [TeeZee/DarkForest-20B-v2.0](https://huggingface.co/TeeZee/DarkForest-20B-v2.0)\n\n### Configuration\n\nThe following YAML configuration was used to produce this model:\n\n```yaml\nslices:\n  - sources:\n      - model: TeeZee/DarkForest-20B-v2.0\n        layer_range: [0, 62]\n      - model: Dust\n        layer_range: [0, 62]\nmerge_method: slerp\nbase_model: TeeZee/DarkForest-20B-v2.0\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\nname: StarDust_20B_v0.2\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_Evillain__StarDust_20B_v0.2)\n\n|             Metric              |Value|\n|---------------------------------|----:|\n|Avg.                             |59.46|\n|AI2 Reasoning Challenge (25-Shot)|61.01|\n|HellaSwag (10-Shot)              |83.76|\n|MMLU (5-Shot)                    |59.29|\n|TruthfulQA (0-shot)              |51.43|\n|Winogrande (5-shot)              |77.27|\n|GSM8k (5-shot)                   |24.03|\n\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 1,
  "downloads": 425,
  "gated": false,
  "private": false,
  "last_modified": "2024-06-07T02:56:37.000Z",
  "created_at": "2024-06-06T19:34:16.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66620f38b177c202745aa778",
  "id": "RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf",
  "modelId": "RichardErkhov/Evillain_-_StarDust_20B_v0.2-gguf",
  "sha": "15698f5d0819a6b57d1fc0bcd0b407bdc2216584",
  "createdAt": "2024-06-06T19:34:16.000Z",
  "lastModified": "2024-06-07T02:56:37.000Z",
  "author": "RichardErkhov",
  "downloads": 425,
  "likes": 1,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}