richarderkhov/openleecher_-_llama_8b_lima_8_kto-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "This model is a fine-tuned version of OpenLeecher/llama_8b_lima_8 on the None dataset. It achieves the following results on the evaluation set:",
    "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\nllama_8b_lima_8_kto - GGUF\n- Model creator: https://huggingface.co/OpenLeecher/\n- Original model: https://huggingface.co/OpenLeecher/llama_8b_lima_8_kto/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [llama_8b_lima_8_kto.Q2_K.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q2_K.gguf) | Q2_K | 2.96GB |\n| [llama_8b_lima_8_kto.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.IQ3_XS.gguf) | IQ3_XS | 3.28GB |\n| [llama_8b_lima_8_kto.IQ3_S.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.IQ3_S.gguf) | IQ3_S | 3.43GB |\n| [llama_8b_lima_8_kto.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q3_K_S.gguf) | Q3_K_S | 3.41GB |\n| [llama_8b_lima_8_kto.IQ3_M.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.IQ3_M.gguf) | IQ3_M | 3.52GB |\n| [llama_8b_lima_8_kto.Q3_K.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q3_K.gguf) | Q3_K | 3.74GB |\n| [llama_8b_lima_8_kto.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q3_K_M.gguf) | Q3_K_M | 3.74GB |\n| [llama_8b_lima_8_kto.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q3_K_L.gguf) | Q3_K_L | 4.03GB |\n| [llama_8b_lima_8_kto.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.IQ4_XS.gguf) | IQ4_XS | 4.18GB |\n| [llama_8b_lima_8_kto.Q4_0.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q4_0.gguf) | Q4_0 | 4.34GB |\n| [llama_8b_lima_8_kto.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.IQ4_NL.gguf) | IQ4_NL | 4.38GB |\n| [llama_8b_lima_8_kto.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q4_K_S.gguf) | Q4_K_S | 4.37GB |\n| [llama_8b_lima_8_kto.Q4_K.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q4_K.gguf) | Q4_K | 4.58GB |\n| [llama_8b_lima_8_kto.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q4_K_M.gguf) | Q4_K_M | 4.58GB |\n| [llama_8b_lima_8_kto.Q4_1.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q4_1.gguf) | Q4_1 | 4.78GB |\n| [llama_8b_lima_8_kto.Q5_0.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q5_0.gguf) | Q5_0 | 5.21GB |\n| [llama_8b_lima_8_kto.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q5_K_S.gguf) | Q5_K_S | 5.21GB |\n| [llama_8b_lima_8_kto.Q5_K.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q5_K.gguf) | Q5_K | 5.34GB |\n| [llama_8b_lima_8_kto.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q5_K_M.gguf) | Q5_K_M | 5.34GB |\n| [llama_8b_lima_8_kto.Q5_1.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q5_1.gguf) | Q5_1 | 5.65GB |\n| [llama_8b_lima_8_kto.Q6_K.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q6_K.gguf) | Q6_K | 6.14GB |\n| [llama_8b_lima_8_kto.Q8_0.gguf](https://huggingface.co/RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf/blob/main/llama_8b_lima_8_kto.Q8_0.gguf) | Q8_0 | 7.95GB |\n\n\n\n\nOriginal model description:\n---\nlibrary_name: transformers\nlicense: llama3\nbase_model: OpenLeecher/llama_8b_lima_8\ntags:\n- trl\n- kto\n- llama-factory\n- generated_from_trainer\nmodel-index:\n- name: llama_8b_lima_8_kto\n  results: []\n---\n\n<!-- This model card has been generated automatically according to the information the Trainer had access to. You\nshould probably proofread and complete it, then remove this comment. -->\n\n# llama_8b_lima_8_kto\n\nThis model is a fine-tuned version of [OpenLeecher/llama_8b_lima_8](https://huggingface.co/OpenLeecher/llama_8b_lima_8) on the None dataset.\nIt achieves the following results on the evaluation set:\n- Loss: 0.1183\n- Rewards/chosen: 5.0516\n- Logps/chosen: -182.1999\n- Rewards/rejected: -5.2649\n- Logps/rejected: -221.3531\n- Rewards/margins: 10.3165\n- Kl: 9.8835\n\n## Model description\n\nMore information needed\n\n## Intended uses & limitations\n\nMore information needed\n\n## Training and evaluation data\n\nMore information needed\n\n## Training procedure\n\n### Training hyperparameters\n\nThe following hyperparameters were used during training:\n- learning_rate: 1e-06\n- train_batch_size: 1\n- eval_batch_size: 1\n- seed: 66\n- distributed_type: multi-GPU\n- num_devices: 2\n- gradient_accumulation_steps: 12\n- total_train_batch_size: 24\n- total_eval_batch_size: 2\n- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08\n- lr_scheduler_type: constant\n- lr_scheduler_warmup_steps: 25\n- num_epochs: 1.0\n\n### Training results\n\n| Training Loss | Epoch  | Step | Validation Loss | Rewards/chosen | Logps/chosen | Rewards/rejected | Logps/rejected | Rewards/margins | Kl     |\n|:-------------:|:------:|:----:|:---------------:|:--------------:|:------------:|:----------------:|:--------------:|:---------------:|:------:|\n| 0.3562        | 0.4995 | 157  | 0.2510          | 2.2688         | -210.0280    | -3.0690          | -199.3941      | 5.3378          | 4.8987 |\n| 0.3412        | 0.9989 | 314  | 0.1183          | 5.0516         | -182.1999    | -5.2649          | -221.3531      | 10.3165         | 9.8835 |\n\n\n### Framework versions\n\n- Transformers 4.45.0\n- Pytorch 2.4.0\n- Datasets 2.21.0\n- Tokenizers 0.20.0\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 473,
  "gated": false,
  "private": false,
  "last_modified": "2024-10-12T00:13:57.000Z",
  "created_at": "2024-10-11T20:14:01.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "67098709258bde9d7ecce05c",
  "id": "RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf",
  "modelId": "RichardErkhov/OpenLeecher_-_llama_8b_lima_8_kto-gguf",
  "sha": "b95dfedcca9cc270372c3cd64298c7b7256b99cc",
  "createdAt": "2024-10-11T20:14:01.000Z",
  "lastModified": "2024-10-12T00:13:57.000Z",
  "author": "RichardErkhov",
  "downloads": 473,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}