Fix TEGroupedMLP pre-backward unshard in fine-grained FSDP hooks for … - #4989
Open
rapatel wants to merge 7 commits into
Open
Fix TEGroupedMLP pre-backward unshard in fine-grained FSDP hooks for …#4989rapatel wants to merge 7 commits into
rapatel wants to merge 7 commits into
Conversation
6 tasks
Contributor
Author
|
/ok to test 0a84107 |
Contributor
Author
|
@cspades PTAL. Thanks! |
Contributor
Author
|
/ok to test 7bd0721 |
rapatel
enabled auto-merge
June 2, 2026 03:31
rapatel
force-pushed
the
ripatel/fsdp_1f1b_pre_bwd_fix_dev
branch
from
June 8, 2026 17:06
7bd0721 to
fa1e49f
Compare
Contributor
Author
|
/ok to test 45cd25c |
Phlip79
approved these changes
Jun 11, 2026
rapatel
force-pushed
the
ripatel/fsdp_1f1b_pre_bwd_fix_dev
branch
2 times, most recently
from
June 25, 2026 18:21
33049d1 to
cc3e139
Compare
Member
|
Had a comment on main branch PR, let's resolve that first: https://github.com/NVIDIA/Megatron-LM/pull/4990/changes#r3478637627 |
rapatel
force-pushed
the
ripatel/fsdp_1f1b_pre_bwd_fix_dev
branch
from
June 26, 2026 23:05
7a04115 to
6256543
Compare
cspades
approved these changes
Jun 29, 2026
…1F1B EP overlap. Signed-off-by: Ritesh Patel <ripatel@nvidia.com>
…_recurse_module_types Signed-off-by: Ritesh Patel <ripatel@nvidia.com>
…, backward]. Signed-off-by: Ritesh Patel <ripatel@nvidia.com>
Signed-off-by: Ritesh Patel <ripatel@nvidia.com>
…e allowlist. Signed-off-by: Ritesh Patel <ripatel@nvidia.com>
rapatel
force-pushed
the
ripatel/fsdp_1f1b_pre_bwd_fix_dev
branch
from
June 30, 2026 00:11
6256543 to
141ae98
Compare
Signed-off-by: Ritesh Patel <ripatel@nvidia.com>
Signed-off-by: Ritesh Patel <ripatel@nvidia.com>
Contributor
|
/ok to test dc9caf9 |
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
4 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.
…1F1B EP overlap.
Main PR - Megatron-LM/pull/4990
What does this PR do ?
Issue tracking
For PRs from open-source community contributors:
Linked issue:
Contribution process
Pre-checks
Code review
Feel free to message or comment @NVIDIA/mcore-oncall to help accelerate your merge into main. The less complex your PR is, the faster it will be approved and merged!
All PRs start as draft. If you open a non-draft PR, it will be automatically converted to draft.
Step 1: Mark PR as "Ready for Review"
.github/CODEOWNERS.Final Review might get declined if these requirements are not fulfilled.
Step 2: Final Review
For PRs that change
megatron/core, once all expert reviewers have approved, theFinal Reviewlabel is applied automatically and final reviewers are assigned.For PRs outside
megatron/core, this step is skipped.Step 3: Approved
Once all required reviewers have approved, the
Approvedlabel is applied automatically.Merge
Any member of mcore-engineers will be able to merge your PR.