This repository hosts a fine-tuned version of the Gemma 3 1B model using Low-Rank Adaptation (LoRA) via KerasNLP on a subset of Medical Question & Answering (Q&A) data.

🚀 Fine-Tuning Details

Base Model: Gemma 3B (1B parameter version)
Framework: TensorFlow / Keras (Saved using .keras format)
Library: KerasNLP
Technique: Parameter-efficient fine-tuning using LoRA
Target Domain: Medical Question Answering
Data Used: A subset of Medical QA data

✅ How to Load the Model

python

from keras.models import load_model

model = load_model("https://huggingface.co/gimmy256/medical-gemma3_1b/resolve/main/medical_gemma3_1b.keras")