make FP4 param gather work with the mixed precisions in NVFP4 recipe

[xrennvidia]

What does this PR do ?

Make --fp4-param-gather work with the mixed precisions in NVFP4 recipe.

Fixes #2284.

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.

[copy-pr-bot]

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.

[github-actions]

This PR has been automatically converted to draft because all PRs must start as drafts.

When you are ready for review, click Ready for Review to begin the review process. This will:

1. Add the oncall reviewer (optional reviewer)
2. Add required review teams based on your changes

See the contribution guide for more details.

[Phlip79]

/claude review

[claude]

Review: Critical bug — TypeError when name is None

The name parameter defaults to None in all modified constructors, but is unconditionally used in string concatenation (e.g., name + ".self_attention"). The primary model construction path through TransformerBlock.build_layer in transformer_block.py does not pass name, so any model using the standard transformer block will crash with:

TypeError: unsupported operand type(s) for +: 'NoneType' and 'str'

This affects all 10+ files where name + "..." concatenation was added.

Suggested fix: Update transformer_block.py:build_layer to pass a name argument (e.g., name=f"decoder.layers.{layer_number}"), and add a defensive guard or make name required to prevent future callers from hitting the same issue.

[Phlip79]

/ok to test c6b5f06

[Phlip79]

/ok to test f326ce3

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/26264704092

[Phlip79]

/ok to test a394695

[svcnvidia-nemo-ci]

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/26270483070