Skip to content

fix: Set num_labels to 1 in causal model initialization for RewardTrainer #5066

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
qgallouedec merged 3 commits into from
Feb 11, 2026
Merged

fix: Set num_labels to 1 in causal model initialization for RewardTrainer #5066

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
f673037
Set num_labels to 1 in model initialization for RewardTrainer
qgallouedec Feb 10, 2026
cc622b2
add comment
qgallouedec Feb 11, 2026
fa16dc3
Merge branch 'main' into fix-reward
qgallouedec Feb 11, 2026
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
1 change: 1 addition & 0 deletions trl/trainer/reward_trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -304,6 +304,7 @@ def __init__(
# Distributed training requires device_map=None ("auto" fails)
if args.distributed_state.distributed_type in ["MULTI_GPU", "DEEPSPEED"]:
model_init_kwargs["device_map"] = None
model_init_kwargs["num_labels"] = 1 # the only output of the model is the reward score
with ignore_seqcls_score_missing_key():
model = create_model_from_path(model, AutoModelForSequenceClassification, **model_init_kwargs)
else:
Expand Down
Loading