richarderkhov/nurtureai_-_neural-chat-11b-v3-1-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "system_input = \"You are a math expert assistant. Your mission is to help users understand and solve various math problems. You should provide step-by-step solutions, explain reasonings and give the correct answer.\" user_input = \"calculate 100 + 520 + 60\" response = generate_response(system_input, user_input) print(response) # expected response \"\"\" To calculate the sum of 100, 520, and 60, we will follow these steps: 1. Add the first two numbers: 100 + 520 2. Add the result from step 1 to the third number: (100 + 520) + 60 Step 1: Add 100 and 520 100 + 520 = 620 Step 2: Add the result from step 1 to the third number (60) (620) + 60 = 680 So, the sum of 100, 520, and 60 is 680. \"\"\" ```",
    "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\nneural-chat-11b-v3-1 - GGUF\n- Model creator: https://huggingface.co/NurtureAI/\n- Original model: https://huggingface.co/NurtureAI/neural-chat-11b-v3-1/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [neural-chat-11b-v3-1.Q2_K.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q2_K.gguf) | Q2_K | 3.73GB |\n| [neural-chat-11b-v3-1.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.IQ3_XS.gguf) | IQ3_XS | 4.14GB |\n| [neural-chat-11b-v3-1.IQ3_S.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.IQ3_S.gguf) | IQ3_S | 4.37GB |\n| [neural-chat-11b-v3-1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q3_K_S.gguf) | Q3_K_S | 4.34GB |\n| [neural-chat-11b-v3-1.IQ3_M.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.IQ3_M.gguf) | IQ3_M | 4.51GB |\n| [neural-chat-11b-v3-1.Q3_K.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q3_K.gguf) | Q3_K | 4.84GB |\n| [neural-chat-11b-v3-1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q3_K_M.gguf) | Q3_K_M | 4.84GB |\n| [neural-chat-11b-v3-1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q3_K_L.gguf) | Q3_K_L | 5.26GB |\n| [neural-chat-11b-v3-1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.IQ4_XS.gguf) | IQ4_XS | 5.43GB |\n| [neural-chat-11b-v3-1.Q4_0.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q4_0.gguf) | Q4_0 | 5.66GB |\n| [neural-chat-11b-v3-1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.IQ4_NL.gguf) | IQ4_NL | 5.72GB |\n| [neural-chat-11b-v3-1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q4_K_S.gguf) | Q4_K_S | 5.7GB |\n| [neural-chat-11b-v3-1.Q4_K.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q4_K.gguf) | Q4_K | 6.02GB |\n| [neural-chat-11b-v3-1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q4_K_M.gguf) | Q4_K_M | 6.02GB |\n| [neural-chat-11b-v3-1.Q4_1.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q4_1.gguf) | Q4_1 | 6.27GB |\n| [neural-chat-11b-v3-1.Q5_0.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q5_0.gguf) | Q5_0 | 6.89GB |\n| [neural-chat-11b-v3-1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q5_K_S.gguf) | Q5_K_S | 6.89GB |\n| [neural-chat-11b-v3-1.Q5_K.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q5_K.gguf) | Q5_K | 7.08GB |\n| [neural-chat-11b-v3-1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q5_K_M.gguf) | Q5_K_M | 7.08GB |\n| [neural-chat-11b-v3-1.Q5_1.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q5_1.gguf) | Q5_1 | 7.51GB |\n| [neural-chat-11b-v3-1.Q6_K.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q6_K.gguf) | Q6_K | 8.2GB |\n| [neural-chat-11b-v3-1.Q8_0.gguf](https://huggingface.co/RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf/blob/main/neural-chat-11b-v3-1.Q8_0.gguf) | Q8_0 | 10.62GB |\n\n\n\n\nOriginal model description:\n---\nlicense: apache-2.0\n---\n\n## 11B\n\n## Original Model Card\n\n## Fine-tuning on Intel Gaudi2\n\nThis model is a fine-tuned model based on [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the open source dataset [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca). Then we align it with DPO algorithm. For more details, you can refer our blog: [The Practice of Supervised Fine-tuning and Direct Preference Optimization on Intel Gaudi2](https://medium.com/@NeuralCompressor/the-practice-of-supervised-finetuning-and-direct-preference-optimization-on-habana-gaudi2-a1197d8a3cd3).\n\n## Model date\nNeural-chat-7b-v3-1 was trained between September and October, 2023.\n\n## Evaluation\n\nWe submit our model to [open_llm_leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard), and the model performance has been **improved significantly** as we see from the average metric of 7 tasks from the leaderboard.\n\n| Model | Average ⬆️| ARC (25-s) ⬆️ | HellaSwag (10-s) ⬆️ | MMLU (5-s) ⬆️| TruthfulQA (MC) (0-s) ⬆️ | Winogrande (5-s) | GSM8K (5-s) | DROP (3-s) |\n| --- | --- | --- | --- | --- | --- | --- | --- | --- |\n|[mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) | 50.32 | 59.58  | 83.31  | 64.16  | 42.15 | 78.37 | 18.12 | 6.14 |\n| [Intel/neural-chat-7b-v3](https://huggingface.co/Intel/neural-chat-7b-v3) | **57.31** | 67.15 | 83.29 | 62.26  | 58.77 | 78.06 | 1.21 | 50.43 |\n| [Intel/neural-chat-7b-v3-1](https://huggingface.co/Intel/neural-chat-7b-v3-1) | **59.06** | 66.21 | 83.64 | 62.37  | 59.65 | 78.14 | 19.56 | 43.84 |\n\n## Training procedure\n\n### Training hyperparameters\n\nThe following hyperparameters were used during training:\n- learning_rate: 1e-04\n- train_batch_size: 1\n- eval_batch_size: 2\n- seed: 42\n- distributed_type: multi-HPU\n- num_devices: 8\n- gradient_accumulation_steps: 8\n- total_train_batch_size: 64\n- total_eval_batch_size: 8\n- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08\n- lr_scheduler_type: cosine\n- lr_scheduler_warmup_ratio: 0.03\n- num_epochs: 2.0\n\n### Training sample code\nHere is the sample code to reproduce the model: [Sample Code](https://github.com/intel/intel-extension-for-transformers/blob/main/intel_extension_for_transformers/neural_chat/examples/finetuning/finetune_neuralchat_v3/README.md).\n\n## Prompt Template\n\n```\n### System:\n{system}\n### User:\n{usr}\n### Assistant:\n\n```\n\n\n## Inference with transformers\n\n```python\nimport transformers\n\n\nmodel_name = 'Intel/neural-chat-7b-v3-1'\nmodel = transformers.AutoModelForCausalLM.from_pretrained(model_name)\ntokenizer = transformers.AutoTokenizer.from_pretrained(model_name)\n\ndef generate_response(system_input, user_input):\n\n    # Format the input using the provided template\n    prompt = f\"### System:\\n{system_input}\\n### User:\\n{user_input}\\n### Assistant:\\n\"\n\n    # Tokenize and encode the prompt\n    inputs = tokenizer.encode(prompt, return_tensors=\"pt\", add_special_tokens=False)\n\n    # Generate a response\n    outputs = model.generate(inputs, max_length=1000, num_return_sequences=1)\n    response = tokenizer.decode(outputs[0], skip_special_tokens=True)\n\n    # Extract only the assistant's response\n    return response.split(\"### Assistant:\\n\")[-1]\n\n\n# Example usage\nsystem_input = \"You are a math expert assistant. Your mission is to help users understand and solve various math problems. You should provide step-by-step solutions, explain reasonings and give the correct answer.\"\nuser_input = \"calculate 100 + 520 + 60\"\nresponse = generate_response(system_input, user_input)\nprint(response)\n\n# expected response\n\"\"\"\nTo calculate the sum of 100, 520, and 60, we will follow these steps:\n\n1. Add the first two numbers: 100 + 520\n2. Add the result from step 1 to the third number: (100 + 520) + 60\n\nStep 1: Add 100 and 520\n100 + 520 = 620\n\nStep 2: Add the result from step 1 to the third number (60)\n(620) + 60 = 680\n\nSo, the sum of 100, 520, and 60 is 680.\n\"\"\"\n\n```\n\n## Ethical Considerations and Limitations\nneural-chat-7b-v3-1 can produce factually incorrect output, and should not be relied on to produce factually accurate information. neural-chat-7b-v3-1 was trained on [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca) based on [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1). Because of the limitations of the pretrained model and the finetuning datasets, it is possible that this model could generate lewd, biased or otherwise offensive outputs.\n\nTherefore, before deploying any applications of neural-chat-7b-v3-1, developers should perform safety testing.\n\n## Disclaimer\n\nThe license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please cosult an attorney before using this model for commercial purposes.\n\n## Organizations developing the model\n\nThe NeuralChat team with members from Intel/DCAI/AISE/AIPT. Core team members: Kaokao Lv, Liang Lv, Chang Wang, Wenxin Zhang, Xuhui Ren, and Haihao Shen.\n\n## Useful links\n* Intel Neural Compressor [link](https://github.com/intel/neural-compressor)\n* Intel Extension for Transformers [link](https://github.com/intel/intel-extension-for-transformers)\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_Intel__neural-chat-7b-v3-1)\n\n| Metric                | Value                     |\n|-----------------------|---------------------------|\n| Avg.                  | 59.06   |\n| ARC (25-shot)         | 66.21          |\n| HellaSwag (10-shot)   | 83.64    |\n| MMLU (5-shot)         | 62.37         |\n| TruthfulQA (0-shot)   | 59.65   |\n| Winogrande (5-shot)   | 78.14   |\n| GSM8K (5-shot)        | 19.56        |\n| DROP (3-shot)         | 43.84         |\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 0,
  "downloads": 114,
  "gated": false,
  "private": false,
  "last_modified": "2024-05-25T16:35:21.000Z",
  "created_at": "2024-05-25T13:39:01.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "6651e9f5d32f6f1f340e857b",
  "id": "RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf",
  "modelId": "RichardErkhov/NurtureAI_-_neural-chat-11b-v3-1-gguf",
  "sha": "eca63e40e8939c981acb7ea298541c3c109c4d1b",
  "createdAt": "2024-05-25T13:39:01.000Z",
  "lastModified": "2024-05-25T16:35:21.000Z",
  "author": "RichardErkhov",
  "downloads": 114,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}