SIGN IN SIGN UP

[Executorch][llama] Allow custom sdpa op replacement pass to leverage attention mask (#10341)

Previously we assumed that the custom sdpa always does causal attention.
This diff adds option to this module swap pass to make custom sdpa leverage
attention mask instead of causal.

Differential Revision: [D73222736](https://our.internmc.facebook.com/intern/diff/D73222736/)
P
pytorchbot committed
8eebbcd108198c02be06ed401535b258a851cac8
Parent: 1f6d3c5
Committed by GitHub <noreply@github.com> on 4/21/2025, 11:12:12 PM