replit / ReplitLM

Inference code and configs for the ReplitLM model family
https://huggingface.co/replit
Apache License 2.0
918 stars 75 forks source link

FlashAttention2 for ReplitLM #29

Open kaushal07wick opened 1 year ago

kaushal07wick commented 1 year ago

I want to add FlashAttention2 to the replit-code-v1 for better performance and efficiency. Please, let me know If i am missing anything. @pirroh @madhavatreplit @amasad