richarderkhov/wassname_-_llama-3-2-1b-sft-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "This model is a fine-tuned version of NousResearch/Llama-3.2-1B on the HuggingFaceH4/ultrachat_200k dataset. It achieves the following results on the evaluation set: See the training yaml https://github.com/wassname/SimPO/blob/main/training_configs/llama-3-2-1b-base-sft.yaml",
    "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-3-2-1b-sft - GGUF\n- Model creator: https://huggingface.co/wassname/\n- Original model: https://huggingface.co/wassname/llama-3-2-1b-sft/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [llama-3-2-1b-sft.Q2_K.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q2_K.gguf) | Q2_K | 0.54GB |\n| [llama-3-2-1b-sft.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.IQ3_XS.gguf) | IQ3_XS | 0.58GB |\n| [llama-3-2-1b-sft.IQ3_S.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.IQ3_S.gguf) | IQ3_S | 0.6GB |\n| [llama-3-2-1b-sft.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q3_K_S.gguf) | Q3_K_S | 0.6GB |\n| [llama-3-2-1b-sft.IQ3_M.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.IQ3_M.gguf) | IQ3_M | 0.61GB |\n| [llama-3-2-1b-sft.Q3_K.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q3_K.gguf) | Q3_K | 0.64GB |\n| [llama-3-2-1b-sft.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q3_K_M.gguf) | Q3_K_M | 0.64GB |\n| [llama-3-2-1b-sft.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q3_K_L.gguf) | Q3_K_L | 0.68GB |\n| [llama-3-2-1b-sft.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.IQ4_XS.gguf) | IQ4_XS | 0.7GB |\n| [llama-3-2-1b-sft.Q4_0.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q4_0.gguf) | Q4_0 | 0.72GB |\n| [llama-3-2-1b-sft.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.IQ4_NL.gguf) | IQ4_NL | 0.72GB |\n| [llama-3-2-1b-sft.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q4_K_S.gguf) | Q4_K_S | 0.72GB |\n| [llama-3-2-1b-sft.Q4_K.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q4_K.gguf) | Q4_K | 0.75GB |\n| [llama-3-2-1b-sft.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q4_K_M.gguf) | Q4_K_M | 0.75GB |\n| [llama-3-2-1b-sft.Q4_1.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q4_1.gguf) | Q4_1 | 0.77GB |\n| [llama-3-2-1b-sft.Q5_0.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q5_0.gguf) | Q5_0 | 0.83GB |\n| [llama-3-2-1b-sft.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q5_K_S.gguf) | Q5_K_S | 0.83GB |\n| [llama-3-2-1b-sft.Q5_K.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q5_K.gguf) | Q5_K | 0.85GB |\n| [llama-3-2-1b-sft.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q5_K_M.gguf) | Q5_K_M | 0.85GB |\n| [llama-3-2-1b-sft.Q5_1.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q5_1.gguf) | Q5_1 | 0.89GB |\n| [llama-3-2-1b-sft.Q6_K.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q6_K.gguf) | Q6_K | 0.95GB |\n| [llama-3-2-1b-sft.Q8_0.gguf](https://huggingface.co/RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf/blob/main/llama-3-2-1b-sft.Q8_0.gguf) | Q8_0 | 1.23GB |\n\n\n\n\nOriginal model description:\n---\nlibrary_name: transformers\nlicense: llama3.2\nbase_model: NousResearch/Llama-3.2-1B\ntags:\n- alignment-handbook\n- generated_from_trainer\ndatasets:\n- HuggingFaceH4/ultrachat_200k\nmodel-index:\n- name: llama-3-2-1b-sft\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-3-2-1b-sft\n\nThis model is a fine-tuned version of [NousResearch/Llama-3.2-1B](https://huggingface.co/NousResearch/Llama-3.2-1B) on the HuggingFaceH4/ultrachat_200k dataset.\nIt achieves the following results on the evaluation set:\n- Loss: 1.2759\n\nSee the training yaml https://github.com/wassname/SimPO/blob/main/training_configs/llama-3-2-1b-base-sft.yaml\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: 2e-05\n- train_batch_size: 8\n- eval_batch_size: 8\n- seed: 42\n- distributed_type: multi-GPU\n- gradient_accumulation_steps: 4\n- total_train_batch_size: 32\n- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08\n- lr_scheduler_type: cosine\n- lr_scheduler_warmup_ratio: 0.1\n- num_epochs: 1\n\n### Training results\n\n| Training Loss | Epoch  | Step | Validation Loss |\n|:-------------:|:------:|:----:|:---------------:|\n| 1.3663        | 0.0534 | 200  | 1.3955          |\n| 1.3413        | 0.1069 | 400  | 1.3722          |\n| 1.365         | 0.1603 | 600  | 1.3632          |\n| 1.33          | 0.2138 | 800  | 1.3532          |\n| 1.3219        | 0.2672 | 1000 | 1.3463          |\n| 1.3355        | 0.3207 | 1200 | 1.3391          |\n| 1.334         | 0.3741 | 1400 | 1.3305          |\n| 1.3183        | 0.4276 | 1600 | 1.3233          |\n| 1.334         | 0.4810 | 1800 | 1.3161          |\n| 1.3013        | 0.5345 | 2000 | 1.3087          |\n| 1.3156        | 0.5879 | 2200 | 1.3016          |\n| 1.3092        | 0.6414 | 2400 | 1.2953          |\n| 1.2518        | 0.6948 | 2600 | 1.2895          |\n| 1.2617        | 0.7483 | 2800 | 1.2846          |\n| 1.3041        | 0.8017 | 3000 | 1.2809          |\n| 1.3102        | 0.8552 | 3200 | 1.2781          |\n| 1.2675        | 0.9086 | 3400 | 1.2765          |\n| 1.2978        | 0.9621 | 3600 | 1.2759          |\n\n\n### Framework versions\n\n- Transformers 4.45.1\n- Pytorch 2.4.1+cu121\n- Datasets 3.0.1\n- Tokenizers 0.20.0\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 181,
  "gated": false,
  "private": false,
  "last_modified": "2024-10-16T16:24:50.000Z",
  "created_at": "2024-10-16T15:43:44.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "670fdf307448cd1945ee8b3d",
  "id": "RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf",
  "modelId": "RichardErkhov/wassname_-_llama-3-2-1b-sft-gguf",
  "sha": "7db97d10ac9b2f3b023788477e254a4cc1e82105",
  "createdAt": "2024-10-16T15:43:44.000Z",
  "lastModified": "2024-10-16T16:24:50.000Z",
  "author": "RichardErkhov",
  "downloads": 181,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}