rare-engineer/MedAssistant-8B

MedAssistant-8B: Medical Reasoning LLM

MedAssistant-8B is an 8 billion parameter Large Language Model (LLM) developed by rare-engineer, specifically fine-tuned for medical reasoning and diagnosis. It is built upon the robust Llama-3.1-8B-Instruct base model and utilizes LoRA (Low-Rank Adaptation) for efficient specialization.

Key Capabilities & Features

• Advanced Medical Reasoning: Designed to assist with complex medical diagnoses by generating detailed, step-by-step explanations using Chain of Thought (CoT).
• Specialized Training Data: Fine-tuned on medical datasets including MedBooks-CoT-18 and MedQA-CoT, enhancing its domain-specific knowledge.
• Performance on Medical Benchmarks: Demonstrates improved performance over its base model on benchmarks like MedQA and MedBullet-5, achieving 69.3% on MedQA and 54.2% test accuracy on MedBullet-5.
• Efficient Fine-tuning: Utilizes LoRA with parameters r=128 and \u03b1=64, making it a resource-efficient specialized model.

When to Use MedAssistant-8B

• Medical Diagnosis Support: Ideal for applications requiring AI assistance in diagnosing medical conditions.
• Medical Question Answering: Suitable for generating comprehensive answers to medical knowledge questions.
• Educational Tools: Can be integrated into platforms for medical students or professionals seeking detailed explanations for clinical scenarios.