Reorder mtp_post_process after attention backward in 1F1B schedule plan#4695
Open
gdengk wants to merge 2 commits into
Open
Reorder mtp_post_process after attention backward in 1F1B schedule plan#4695gdengk wants to merge 2 commits into
gdengk wants to merge 2 commits into
Conversation
Move mtp_post_process.forward out of the moe_combine fp8 context block so it runs after the attn backward, improving overlap between the forward mtp_post_process and the backward attn compute. (cherry picked from commit 2e7fdc3)
svcnvidia-nemo-ci
added
Final Review
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Ports #4430 from dev to main.
Moves mtp_post_process.forward after attention backward in the 1F1B schedule plan so the forward post-process no longer runs inside the moe_combine FP8 context block before attention backward.
Original dev PR: #4430
Original dev commit: 2e7fdc3