Instruct-tune LLaMA on consumer hardware

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

A Unified Library for Parameter-Efficient and Modular Transfer Learning

A playbook for systematically maximizing the performance of deep learning models.