transformers
Fix DogeForCausalLM attending to future tokens under sdpa
#48770
Open

Fix DogeForCausalLM attending to future tokens under sdpa #48770

Omarinho12345
Omarinho12345 fix: always materialize causal mask for Doge (sdpa future-token leak)
4c9d2410
github-actions
github-actions

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
Milestone