PowerRetention: a drop-in replacement for FlashAttention in LLMs(github.com)
github.com
PowerRetention: a drop-in replacement for FlashAttention in LLMs
https://github.com/m-a-n-i-f-e-s-t/retention
2 comments
Here’s an article I found from Manifest AI, the company behind Power Retention: https://manifestai.com/articles/what-is-power-retention/
I am one of the authors, happy to answer any questions!