-
Notifications
You must be signed in to change notification settings - Fork 561
FA3 #3623
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
base: main
Are you sure you want to change the base?
FA3 #3623
Conversation
zhaochaoxing
commented
Jun 9, 2025
failed on second round chat |
The bug has been resolved. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
I'd like to hold this PR for a while. |
@@ -202,7 +202,8 @@ def flatten_kv_cache(k_caches: Tensor, | |||
k_scales_zeros: Tensor = None, | |||
v_scales_zeros: Tensor = None, | |||
quant_policy: Literal[0, 4, 8] = 0, | |||
kv_layout: str = 'bshd'): | |||
kv_layout: str = 'bshd', | |||
flatten_kv_layout: str = 'bhsd'): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since output is 3d continuous batching tensors, I think 'hsd' is better.