transformers
In Llama4 fix wrongly inverted causal attention mask when using SDPA implementation
#38094

Merged

In Llama4 fix wrongly inverted causal attention mask when using SDPA implementation #38094

ArthurZucker merged 2 commits into huggingface:main from sogartar:fix-llama4-sdpa-causal-attention-mask

In Llama4 fix wrongly inverted causal attention mask when using SDPA …

a33a1836

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

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

github-actions requested a review from

ArthurZucker 1 year ago

github-actions requested a review from

Rocketknight1 1 year ago

ArthurZucker approved these changes on 2025-05-20

Merge branch 'main' into fix-llama4-sdpa-causal-attention-mask

f7666100

ArthurZucker merged 2ad152f8 into main 1 year ago

Reviewers

ArthurZucker

Rocketknight1

Assignees

No one assigned

Labels

None yet

Milestone

No milestone