shafire/QuantumAI
1
1---2language: en3tags:4 - text-generation5 - transformers6 - conversational7 - quantum-math8 - PEFT9 - Safetensors10 - AutoTrain11license: other12datasets: conversational-dataset13model-index:14 - name: Zero LLM Quantum AI15 results:16 - task:17 type: text-generation18 dataset:19 name: conversational-dataset20 type: text21 metrics:22 - name: Training Loss23 type: loss24 value: 1.7425---26 27# **QuantumAI: Zero LLM Quantum AI Model**28 29**Zero Quantum AI** is an LLM that tries to bypass needing quantum computing using theoretical interdimensional mathematics, quantum math, and the **Mathematical Probability of Goodness**. Developed by **TalkToAi.org** and **ResearchForum.Online**, this model leverages cutting-edge AI frameworks to redefine conversational AI, ensuring deep, ethical decision-making capabilities. The model is fine-tuned on **Meta-Llama-3.1-8B-Instruct** and trained via **AutoTrain** to optimize conversational tasks, dialogue generation, and inference.30 3132 33## **Model Information**34 35- **Base Model**: `meta-llama/Meta-Llama-3.1-8B`36- **Fine-tuned Model**: `meta-llama/Meta-Llama-3.1-8B-Instruct`37- **Training Framework**: `AutoTrain`38- **Training Data**: Conversational and text-generation focused dataset39 40### **Tech Stack**41 42- Transformers43- PEFT (Parameter-Efficient Fine-Tuning)44- TensorBoard (for logging and metrics)45- Safetensors46 47### **Usage Types**48 49- Interactive dialogue50- Text generation51 52### **Key Features**53 54- **Quantum Mathematics & Interdimensional Calculations**: Utilizes quantum principles to predict user intent and generate insightful responses.55- **Mathematical Probability of Goodness**: All responses are ethically aligned using a mathematical framework, ensuring positive interactions.56- **Efficient Inference**: Supports **4-bit quantization** for faster and resource-efficient deployment.57 58## **Installation and Usage**59 60To use the model in your Python code:61 62```python63from transformers import AutoModelForCausalLM, AutoTokenizer64 65model_path = "PATH_TO_THIS_REPO"66 67tokenizer = AutoTokenizer.from_pretrained(model_path)68model = AutoModelForCausalLM.from_pretrained(69 model_path,70 device_map="auto",71 torch_dtype='auto'72).eval()73 74# Example usage75messages = [76 {"role": "user", "content": "hi"}77]78 79input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')80output_ids = model.generate(input_ids.to('cuda'))81response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)82 83# Output84print(response)85 86## **Inference API**87 88This model is not yet deployed to the Hugging Face Inference API. However, you can deploy it to **Inference Endpoints** for dedicated, serverless inference.89 90## **Training Process**91 92The **Zero Quantum AI** model was trained using **AutoTrain** with the following configuration:93 94- **Hardware**: CUDA 12.195- **Training Precision**: Mixed FP1696- **Batch Size**: 297- **Learning Rate**: 3e-0598- **Epochs**: 599- **Optimizer**: AdamW100- **PEFT**: Enabled (LoRA with lora_r=16, lora_alpha=32)101- **Quantization**: Int4 for efficient deployment102- **Scheduler**: Linear with warmup103- **Gradient Accumulation**: 4 steps104- **Max Sequence Length**: 2048 tokens105 106## **Training Metrics**107 108Monitored using **TensorBoard**, with key training metrics:109 110- **Training Loss**: 1.74111- **Learning Rate**: Adjusted per epoch, starting at 3e-05.112 113## **Model Features**114 115- **Text Generation**: Handles various types of user queries and provides coherent, contextually aware responses.116- **Conversational AI**: Optimized specifically for generating interactive dialogues.117- **Efficient Inference**: Supports Int4 quantization for faster, resource-friendly deployment.118 119## **License**120 121This model is governed under a custom license. Please refer to [QuantumAI License](https://huggingface.co/shafire/QuantumAI) for details, in compliance with **Meta-Llama 3.1 License**.122