Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

remove deprecated attn_fn #180

Closed
wants to merge 1 commit into from
Closed

remove deprecated attn_fn #180

wants to merge 1 commit into from

Conversation

samsja
Copy link
Collaborator

@samsja samsja commented Dec 20, 2024

No description provided.

@Jackmin801
Copy link
Member

If its not too hard to recall, why did we deprecate this?
What do you think about undeprecating it? I think it can be pretty useful to swap attn impls so id prefer to keep it

Copy link
Member

@Jackmin801 Jackmin801 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hrmm I dont think we should deprecate this

@samsja
Copy link
Collaborator Author

samsja commented Dec 23, 2024

If its not too hard to recall, why did we deprecate this? What do you think about undeprecating it? I think it can be pretty useful to swap attn impls so id prefer to keep it

as of today we only have flex_attention or sdpa. The flash attention one has been removed. To dispatch to flex or sdpa is based on the sequence_pakcing parameters

@samsja
Copy link
Collaborator Author

samsja commented Jan 5, 2025

closing as we decided to use this flag

@Jackmin801 Jackmin801 closed this Jan 6, 2025
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.

2 participants