Skip to content

Document causal mask alignment in scaled_dot_product_attention #2967

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.

Sign up for GitHub

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

Jump to bottom
Open
Anri-Lombard wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Document causal mask alignment in scaled_dot_product_attention #2967

Anri-Lombard wants to merge 1 commit into from

Conversation

@Anri-Lombard
Copy link

@Anri-Lombard Anri-Lombard commented Jan 4, 2026

Summary

  • Document that MLX's mask="causal" uses lower-right alignment
  • Clarify the difference from PyTorch's default is_causal=True (upper-left)

When T_q != T_kv, this distinction matters:

  • MLX (lower-right): Last query aligns with last key
  • PyTorch default (upper-left): First query aligns with first key

References:

Relates to #2835

@Anri-Lombard
Document causal mask alignment in scaled_dot_product_attention
c55b3f5 
Clarify that MLX uses lower-right alignment for causal masks when
T_q != T_kv, which differs from PyTorch's default upper-left alignment.

Relates to ml-explore#2835
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@Anri-Lombard