richarderkhov/hemanth-thunder_-_tamil-mistral-7b-v0.1-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "loss_graph.png",
    "summary": "The Tamil-Mistral-7B-v0.1 Large Language Model (LLM) is a pre-trained generative text model trained at the top of mistral base model 7 billion parameters. This is extends version of tokenization capability by increasing tamil tokens by 20k. Additionally, it was Pretrained on 1.19 million Tamil documents sourced from madlad-400 (Tamil) MADLAD-400 (Multilingual Audited Dataset: Low-resource And Document-level). pretraining time: 145 hours (GPU NVIDIA RTX A6000 48GB)",
    "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\nTamil-Mistral-7B-v0.1 - GGUF\n- Model creator: https://huggingface.co/Hemanth-thunder/\n- Original model: https://huggingface.co/Hemanth-thunder/Tamil-Mistral-7B-v0.1/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Tamil-Mistral-7B-v0.1.Q2_K.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q2_K.gguf) | Q2_K | 2.61GB |\n| [Tamil-Mistral-7B-v0.1.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.IQ3_XS.gguf) | IQ3_XS | 2.9GB |\n| [Tamil-Mistral-7B-v0.1.IQ3_S.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.IQ3_S.gguf) | IQ3_S | 3.05GB |\n| [Tamil-Mistral-7B-v0.1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q3_K_S.gguf) | Q3_K_S | 3.04GB |\n| [Tamil-Mistral-7B-v0.1.IQ3_M.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.IQ3_M.gguf) | IQ3_M | 3.15GB |\n| [Tamil-Mistral-7B-v0.1.Q3_K.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q3_K.gguf) | Q3_K | 3.37GB |\n| [Tamil-Mistral-7B-v0.1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q3_K_M.gguf) | Q3_K_M | 3.37GB |\n| [Tamil-Mistral-7B-v0.1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q3_K_L.gguf) | Q3_K_L | 3.65GB |\n| [Tamil-Mistral-7B-v0.1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.IQ4_XS.gguf) | IQ4_XS | 3.77GB |\n| [Tamil-Mistral-7B-v0.1.Q4_0.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q4_0.gguf) | Q4_0 | 3.93GB |\n| [Tamil-Mistral-7B-v0.1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.IQ4_NL.gguf) | IQ4_NL | 3.97GB |\n| [Tamil-Mistral-7B-v0.1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q4_K_S.gguf) | Q4_K_S | 3.95GB |\n| [Tamil-Mistral-7B-v0.1.Q4_K.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q4_K.gguf) | Q4_K | 4.17GB |\n| [Tamil-Mistral-7B-v0.1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q4_K_M.gguf) | Q4_K_M | 4.17GB |\n| [Tamil-Mistral-7B-v0.1.Q4_1.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q4_1.gguf) | Q4_1 | 4.34GB |\n| [Tamil-Mistral-7B-v0.1.Q5_0.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q5_0.gguf) | Q5_0 | 4.76GB |\n| [Tamil-Mistral-7B-v0.1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q5_K_S.gguf) | Q5_K_S | 4.76GB |\n| [Tamil-Mistral-7B-v0.1.Q5_K.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q5_K.gguf) | Q5_K | 4.89GB |\n| [Tamil-Mistral-7B-v0.1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q5_K_M.gguf) | Q5_K_M | 4.89GB |\n| [Tamil-Mistral-7B-v0.1.Q5_1.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q5_1.gguf) | Q5_1 | 5.18GB |\n| [Tamil-Mistral-7B-v0.1.Q6_K.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q6_K.gguf) | Q6_K | 5.65GB |\n| [Tamil-Mistral-7B-v0.1.Q8_0.gguf](https://huggingface.co/RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf/blob/main/Tamil-Mistral-7B-v0.1.Q8_0.gguf) | Q8_0 | 7.32GB |\n\n\n\n\nOriginal model description:\n---\nlanguage:\n- ta\nlicense: apache-2.0\ntags:\n- pretrained\ndatasets:\n- Hemanth-thunder/tamil-madlad-400\npipeline_tag: text-generation\ninference:\n  parameters:\n    temperature: 0.7\n    repetition_penalty: 1.15\n---\n# Model Card for Tamil-Mistral-7B-v0.1\n\nThe Tamil-Mistral-7B-v0.1 Large Language Model (LLM) is a pre-trained generative text model trained at the top of mistral base model 7 billion parameters. This is extends version of tokenization capability by increasing tamil tokens by 20k. \nAdditionally, it was Pretrained on 1.19 million Tamil documents sourced from madlad-400 (Tamil) [MADLAD-400 (Multilingual Audited Dataset: Low-resource And Document-level)](https://arxiv.org/abs/2309.04662).\n\npretraining time: 145 hours (GPU NVIDIA RTX A6000 48GB)\n## Mistral model details\n\nFor full details of this model please read our [paper](https://arxiv.org/abs/2310.06825) and [release blog post](https://mistral.ai/news/announcing-mistral-7b/).\n\n## Model Architecture\n\nMistral-7B-v0.1 is a transformer model, with the following architecture choices:\n- Grouped-Query Attention\n- Sliding-Window Attention\n- Byte-fallback BPE tokenizer\n\n[Kaggle Demo](https://www.kaggle.com/code/hemanthkumar21/tamil-mistral-7b-v0-1-demo/)\n\n#### Running the model on a GPU 16GB \n\n\n```python\nimport torch\nfrom transformers import (AutoModelForCausalLM,AutoTokenizer,TextStreamer,pipeline)\nmodel = AutoModelForCausalLM.from_pretrained(\"Hemanth-thunder/Tamil-Mistral-7B-v0.1\",device_map=\"auto\")\ntokenizer = AutoTokenizer.from_pretrained(\"Hemanth-thunder/Tamil-Mistral-7B-v0.1\",add_prefix_space=True)\ntokenizer.pad_token = tokenizer.eos_token\ntokenizer.padding_side = \"right\"\nstreamer = TextStreamer(tokenizer)\npipe = pipeline(\"text-generation\" ,model=model, tokenizer=tokenizer ,do_sample=True, repetition_penalty=1.15,top_p=0.95,streamer=streamer)\npipe(\"ஐபிஎல் தொடரில் மும்பை இந்தியன்ஸ் அணி \",max_length=50)\n```\n\n```generated_text\n\nஐபிஎல் தொடரில் மும்பை இந்தியன்ஸ் அணி -3வது இடத்திற்கு முன்னேறி இருக்கிறது, இதனால் பிளே ஆஃப் வாய்ப்பை உறுதி செய்ய வேண்டும்.\nஇன்னும் 11 புள்ளிகள் மட்டுமே மீதமுள்ளது.சென்னை சூப்பர் கிங்சுக்கு 12 புள்ளிகளில் உள்ளது.\nஅதன் கடைசி லீக் போட்டி ஜூன் 23-ம் தேதி சென்னையில் நடைபெறுகிறது.\n\n```\n# Loss\n\n<!-- Provide a quick summary of what the model is/does. -->\n![row01](loss_graph.png)\n\n\n## Troubleshooting\n\n- If you see the following error:\n```\nKeyError: 'mistral'\n```\n- Or:\n```\nNotImplementedError: Cannot copy out of meta tensor; no data!\n```\n\nEnsure you are utilizing a stable version of Transformers, 4.34.0 or newer.\n\n## Notice\n\nMistral 7B is a pretrained base model and therefore does not have any moderation mechanisms.\n\n# How to Cite\n\n```bibtext\n@misc{Tamil-Mistral-7B-v0.1, \n      url={[https://huggingface.co/Hemanth-thunder/Tamil-Mistral-7B-v0.1]https://huggingface.co/Hemanth-thunder/Tamil-Mistral-7B-v0.1)}, \n      title={Tamil-Mistral-7B-v0.1}, \n      author={\"hemanth kumar\"}\n}\n```\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "arxiv:2309.04662",
    "arxiv:2310.06825",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 208,
  "gated": false,
  "private": false,
  "last_modified": "2024-06-26T05:38:51.000Z",
  "created_at": "2024-06-26T01:51:08.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "667b740c81adc76dc71dae61",
  "id": "RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf",
  "modelId": "RichardErkhov/Hemanth-thunder_-_Tamil-Mistral-7B-v0.1-gguf",
  "sha": "1f6f04209286b04ac9cade8f65d0a0b900daf4e8",
  "createdAt": "2024-06-26T01:51:08.000Z",
  "lastModified": "2024-06-26T05:38:51.000Z",
  "author": "RichardErkhov",
  "downloads": 208,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}