Fix T5Attention shape mismatch under Tensor Parallelism #45109
Fix T5Attention shape mismatch under Tensor Parallelism
e9f299e1
vasqu
approved these changes
on 2026-03-30
Run make fix-repo: propagate TP view fix to copied models
d4f43580
Refactor view() calls to use shape tuples per review
8680ba54
vasqu
approved these changes
on 2026-03-31
Chain view() calls per reviewer suggestion
89d64b49
vasqu
enabled auto-merge 46 days ago
vasqu
merged
d55f0350
into main 46 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub