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.

Sign up for GitHub

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

Jump to bottom

why generate use flash-attn is slower? #239

Open
dshnightmare opened this issue Mar 11, 2024 · 2 comments
Open

why generate use flash-attn is slower? #239

dshnightmare opened this issue Mar 11, 2024 · 2 comments
Assignees
@wuxibin89

Comments

@dshnightmare
Copy link

single node deepspeed launch
@hijkzzz
Copy link
Collaborator

hijkzzz commented Mar 12, 2024

Interestingly,flash attention is mainly used to speed up the training phase

@dshnightmare
Copy link
Author

In my test, when diable flash attention, generation is nearly 2x faster....

@wuxibin89 wuxibin89 self-assigned this Mar 15, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@wuxibin89 wuxibin89

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@dshnightmare @wuxibin89 @hijkzzz