This is a fine-tuned version of the Falcon-7B-Instruct model, adapted for generating medical text related to common diseases. The model has been fine-tuned using LoRA (Low-Rank Adaptation) on a dataset of medical texts.
Model Details
Base Model: tiiuae/falcon-7b-instruct
Fine-Tuning Method: LoRA (Low-Rank Adaptation)
Quantization: 4-bit (using bitsandbytes)
Training Dataset: Medical text data (common diseases)
Training Framework: PyTorch with Hugging Face Transformers
Fine-Tuning Duration: 3 epochs
Learning Rate: 1e-3
Batch Size: 2 (per device)
Usage
You can use this model for generating medical text or answering questions related to common diseases.
