This repo contains a low-rank adapter for LLaMA-7b model fit on the Stanford Alpaca dataset. Also modified for enhance performance with 8 epoch training.