This repository houses a fork of togethercomputer/LLaMA-2-7B-32K's modeling_flash_llama.py, with a fix for padding of attention weights merged into it.