Skip to content

[Dev] fix(layer_wise): tag MTP-stage word_embeddings as is_embedding_or_output_parameter #5180

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
Wohox wants to merge 1 commit into from
Closed

[Dev] fix(layer_wise): tag MTP-stage word_embeddings as is_embedding_or_output_parameter #5180

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions megatron/core/models/common/language_module/language_module.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
# Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
import logging
import os
from typing import Optional, Tuple
Expand Down Expand Up @@ -202,7 +202,8 @@ def setup_embeddings_and_output_layer(self) -> None:

# Mark embedding and output layer for decoupled_lr and other features.
# This is the original Megatron attribute used by decoupled_lr, Muon, FSDP, etc.
if self.pre_process and hasattr(self, 'embedding'):
# Include MTP-stage embedding too: it is a duplicated copy of the pre_process embedding
if (self.pre_process or getattr(self, 'mtp_process', False)) and hasattr(self, 'embedding'):
self.embedding.word_embeddings.weight.is_embedding_or_output_parameter = True
if (
self.post_process
Expand Down