transformers
[performance_optim] reduce frequency of declaring attention_mask in Ascend NPU flash attention
#38278
Merged

[performance_optim] reduce frequency of declaring attention_mask in Ascend NPU flash attention #38278

FightingZhen
FightingZhen FightingZhen force pushed from 5fadf99e to a43c506a 312 days ago
FightingZhen
FightingZhen FightingZhen force pushed from a43c506a to 14928611 312 days ago
FightingZhen [performance_optim] reduce frequency of declaring attention_mask in A…
5b31bb0e
FightingZhen FightingZhen force pushed from 14928611 to 5b31bb0e 312 days ago
SunMarc
SunMarc commented on 2025-05-22
SunMarc SunMarc requested a review from ArthurZucker ArthurZucker 311 days ago
FightingZhen
HuggingFaceDocBuilderDev
CurryRice233
FightingZhen
ArthurZucker
ArthurZucker approved these changes on 2025-05-23
ArthurZucker ArthurZucker merged 3c289e21 into main 310 days ago
FightingZhen FightingZhen deleted the optim_fa2_perf branch 228 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone