PR #37285 Fixing flex attention for torch=2.6.0

Fixing flex attention for torch=2.6.0 #37285

ArthurZucker merged 6 commits into huggingface:main from salmanmohammadi:flex_2_6_fix

adding compile kwarg for torch 2.6

a026a6ae

fixing dynamic

0d8ba723

github-actions marked this pull request as draft 1 year ago

salmanmohammadi marked this pull request as ready for review 1 year ago

github-actions requested a review from

MekkCyber 1 year ago

github-actions requested a review from

SunMarc 1 year ago

vasqu commented on 2025-04-04

addressing comment

f13dc1bd

typo

c636e591

ArthurZucker approved these changes on 2025-04-07

Update src/transformers/integrations/flex_attention.py

48f8964d

Merge branch 'main' into flex_2_6_fix

9efb0d46

ArthurZucker added for patch

ArthurZucker merged 794fde7b into main 1 year ago

Reviewers

ArthurZucker

vasqu

MekkCyber

SunMarc

Assignees

No one assigned

Labels

for patch

Milestone

No milestone