Skip to content

Add argument for setting inference logits dtype#4584

Draft
santhnm2 wants to merge 3 commits intoNVIDIA:mainfrom
santhnm2:inference_logits_dtype
Draft

Add argument for setting inference logits dtype#4584
santhnm2 wants to merge 3 commits intoNVIDIA:mainfrom
santhnm2:inference_logits_dtype

Conversation

@santhnm2
Copy link
Copy Markdown
Contributor

@santhnm2 santhnm2 commented May 1, 2026

What does this PR do ?

The inference logits dtype previously relied implicitly on Float16Module defaulting fp32_output to True to ensure fp32 logits. Also, the pipeline parallel path converted logits to params_dtype for buffer dtype consistency. Now everything is standardized around --inference-logits-dtype which defaults to fp32 and is used explicitly.

Issue tracking

For PRs from open-source community contributors:

  • New features: a linked issue is required. Please open a feature request and reference it here before submitting the PR.
  • Small updates (bug fixes, minor improvements): a linked issue is recommended and will accelerate the PR review process.

Linked issue:

Contribution process

Pre-checks

  • I have added relevant unit tests
  • I have added relevant functional tests
  • I have added proper typing to my code Typing guidelines
  • I have added relevant documentation
  • I have run the autoformatter.sh on my PR

Code review

Feel free to message or comment the @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"

  1. When your PR is ready, click Ready for Review.
  2. An oncall reviewer is auto-assigned and expert reviewers are notified based on your changes.
    • Some PRs may jump straight to step 2. This is determined by .github/CODEOWNERS.

⚠️ Only mark as ready once merge-conflicts are resolved and the CI is passing.
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, the Final Review label 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 Approved label is applied automatically.

Merge

Any member of mcore-engineers will be able to merge your PR.

For MRs into `dev` branch The proposed review process for `dev` branch is under active discussion.

MRs are mergable after one approval by either eharper@nvidia.com or zijiey@nvidia.com.

Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>
@copy-pr-bot
Copy link
Copy Markdown

copy-pr-bot Bot commented May 1, 2026

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@santhnm2
Copy link
Copy Markdown
Contributor Author

santhnm2 commented May 1, 2026

/ok to test b7542b7

Signed-off-by: Keshav Santhanam <ksanthanam@nvidia.com>
@santhnm2
Copy link
Copy Markdown
Contributor Author

santhnm2 commented May 1, 2026

/ok to test 665ec48

@santhnm2 santhnm2 changed the title Add CLI arg for setting inference logits dtype Add argument for setting inference logits dtype May 1, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant