Skip to content

Revert "[recipe] feat: enable THD packing by default for Qwen3.5-VL finetune" #3894

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

Merged
cuichenx merged 1 commit into from
May 19, 2026
+1 −4
Merged

Revert "[recipe] feat: enable THD packing by default for Qwen3.5-VL finetune" #3894

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: 1 addition & 4 deletions src/megatron/bridge/recipes/qwen_vl/qwen35_vl.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -125,10 +125,7 @@ def _qwen35_vl_apply_common(
# Dataset configuration
cfg.dataset.seq_length = 4096
cfg.dataset.hf_processor_path = hf_path
# Enable THD packing by default for SFT/PEFT: the step function builds
# cu_seqlens so attention skips padding FLOPs on short samples (especially
# when PP/EP forces micro-batches to pad to seq_length).
cfg.dataset.pack_sequences_in_batch = True
cfg.dataset.pack_sequences_in_batch = False

# DDP settings
cfg.ddp.overlap_grad_reduce = False
Expand Down
Loading