Skip to content

Conversation

@AdityaKulshrestha
Copy link

Hi team,

I have added fix for I guess an older method name in the docstring of PagedAttention. It has been using flash_atten_varlen, I have updated it to flash_attn_varlen_func.

The parameters for enabling causal masking is referred as is_cusal. I am not sure if it was also done on purpose or just a typo since there are multiple instances of the same.

Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant