Run LLaMA (and Stanford-Alpaca) inference on Apple Silicon GPUs.
The open source implementation of ChatGPT, Alpaca, Vicuna and RLHF Pipe...