fix(mtp): support multi-head MTP loss logging (mtp-num-layers > 1)#2132
Draft
ZiyiTsang wants to merge 1 commit into
Draft
fix(mtp): support multi-head MTP loss logging (mtp-num-layers > 1)#2132ZiyiTsang wants to merge 1 commit into
ZiyiTsang wants to merge 1 commit into
Conversation
slime's per-step MTP-loss logging called .item() on tracker["values"], which is a [num_mtp_layers] vector from Megatron's MTPLossLoggingHelper. This crashes for any multi-head MTP model (--mtp-num-layers > 1) under --enable-mtp-training with: RuntimeError: a Tensor with N elements cannot be converted to Scalar. Align with Megatron's track_mtp_metrics: keep the per-layer vector and log one scalar per layer plus a summed total. Co-Authored-By: Claude <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Quick Fix for issue #2131.
Support multi-head MTP loss logging (mtp-num-layers > 1).