fix(megatron-fsdp): preserve non-meta tensors during meta device materialization

[xuwchen]

What does this PR do ?

During meta device initialization, the materialization path calls m.to_empty(device, recurse=False) before reset_parameters(). to_empty() unconditionally replaces all tensors (both parameters and buffers) in the module with uninitialized memory, regardless of whether they are actually on the meta device.

This becomes a problem for MoE when --moe-router-enable-expert-bias is enabled: expert_bias is registered with an explicit device=torch.cuda.current_device() [link], so it bypasses the meta context and is correctly initialized to zeros on GPU. But to_empty() clobbers it with uninitialized memory (NaN), and reset_parameters() only reinitializes parameters, not buffers, so expert_bias stays NaN for the entire training run.

The fix here replaces m.to_empty() with a targeted helper that only materializes is_meta=True tensors, leaving non-meta buffers (like expert_bias) untouched.

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.

[shjwudp]

LGTM

[cspades]

Can someone give me a short backstory for why this is even needed (for MoE)? Why do we care about values prior to module.reset_parameters()? 👀

[xuwchen]

Can someone give me a short backstory for why this is even needed (for MoE)? Why do we care about values prior to module.reset_parameters()? 👀

During meta device initialization, the materialization path calls m.to_empty(device, recurse=False) before reset_parameters(). to_empty() unconditionally replaces all tensors (both parameters and buffers) in the module with uninitialized memory, regardless of whether they are actually on the meta device.

This becomes a problem for MoE when --moe-router-enable-expert-bias is enabled: expert_bias is registered with an explicit device=torch.cuda.current_device() [link], so it bypasses the meta context and is correctly initialized to zeros on GPU. But to_empty() clobbers it with uninitialized memory (NaN), and reset_parameters() only reinitializes parameters, not buffers, so expert_bias stays NaN for the entire training run.

The fix here replaces m.to_empty() with a targeted helper that only materializes is_meta=True tensors, leaving non-meta buffers (like expert_bias) untouched.

[cspades]

/ok to test f13f04b

[Phlip79]

/ok to test 626235a