Skip to content
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

Update TE version for support of pad_between_seqs=True #12174

Open
cyanguwa opened this issue Feb 13, 2025 · 0 comments
Open

Update TE version for support of pad_between_seqs=True #12174

cyanguwa opened this issue Feb 13, 2025 · 0 comments
Labels
bug Something isn't working

Comments

@cyanguwa
Copy link
Contributor

Describe the bug

Currently, NeMo 25.02rc2 container uses TE 1.14.0+87fbe81, which blocks certain attention use cases for pad_between_seqs=True when qkv_format=thd. Please include this PR in the next container build:
NVIDIA/TransformerEngine#1387

In particular, these lines are the fix:
https://github.com/NVIDIA/TransformerEngine/blob/f0d22ca12f574233053da20516997e45d99eb65c/transformer_engine/pytorch/attention.py#L7457-L7463
@cyanguwa cyanguwa added the bug Something isn't working label Feb 13, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@cyanguwa