graliuce/Qwen2.5-3B-Instruct_MedMCQA.20.01_terminate_aug_balanced_1.0e-5
This model is a 3.1 billion parameter instruction-tuned causal language model, fine-tuned from Qwen/Qwen2.5-3B-Instruct by graliuce. It has been specifically adapted for medical question answering tasks using the MedMCQA.20.01_terminate_aug_balanced dataset. The model is optimized for specialized medical domain understanding and response generation.
Loading preview...
Model Overview
This model, developed by graliuce, is a specialized fine-tuned version of the Qwen2.5-3B-Instruct architecture, featuring 3.1 billion parameters. It has been specifically adapted for the medical domain through training on the graliuce/MedMCQA.20.01_terminate_aug_balanced dataset.
Key Capabilities
- Specialized Medical QA: Designed to answer questions within the medical domain.
- Instruction Following: Inherits instruction-following capabilities from its base Qwen2.5-3B-Instruct model.
- Fine-tuned Performance: Utilizes Supervised Fine-Tuning (SFT) with the TRL library for domain adaptation.
Training Details
The model was trained using the TRL (Transformer Reinforcement Learning) library, leveraging a balanced and augmented version of the MedMCQA dataset. This targeted training aims to enhance its performance on medical multiple-choice question answering. The training process involved specific versions of frameworks including TRL 0.17.0, Transformers 4.52.3, Pytorch 2.6.0, Datasets 3.2.0, and Tokenizers 0.21.4.
Use Cases
This model is particularly suitable for applications requiring accurate and contextually relevant responses to medical questions, such as educational tools for medical students, preliminary diagnostic support systems, or information retrieval in healthcare contexts.