deepseek: fix conversion script + adopt callbacks to an API change in Megatron-Bridge#868
Merged
ShriyaRishab merged 26 commits intomlcommons:masterfrom Apr 16, 2026
Conversation
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
Contributor
|
@denys-fridman can you provide a description of this change and explain if it impacts convergence/accuracy/RCPs? |
Contributor
Author
|
@ShriyaRishab Added the description. It's still a Draft and is in the process of testing. |
Contributor
|
@denys-fridman is this ready? |
…R_DECAY_STEPS=11996
…on.bridge Replace custom TrainingCallback protocol and CallbackManager with the new Callback/CallbackContext API from megatron.bridge.training.callbacks. Rename on_validation_start/end -> on_eval_start/end and on_train_batch_end -> on_train_step_end. Pass callbacks list directly to pretrain() instead of using register_callback/install_callbacks. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
ConfigContainer uses 'validate' not 'validation' as the attribute name. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
cfg.validate is a method on ConfigContainer; eval_interval lives on cfg.train. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
suexu1025
approved these changes
Apr 16, 2026
suexu1025
left a comment
There was a problem hiding this comment.
LGTM, nit: would that be possible to commit the patches in the repro instead of manual patching?
Contributor
|
@denys-fridman - can you please create a separate PR committing the patches to this repo as requested above? |
ShriyaRishab
approved these changes
Apr 16, 2026
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 subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
This MR:
None of the changes affect the convergence