fix pp memory issue (#6344)

Co-authored-by: Tong Li <tong.li35271158@gmail.com>
This commit is contained in:
Tong Li 2025-06-11 17:54:18 +08:00 committed by GitHub
parent bb6f5d98fc
commit 8992def757
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -345,7 +345,7 @@ class GRPOConsumer(BaseConsumer):
criterion=_criterion, criterion=_criterion,
optimizer=self.optimizer, optimizer=self.optimizer,
return_loss=True, return_loss=True,
return_outputs=True, return_outputs=False,
) )
loss = policy_model_outputs["loss"] loss = policy_model_outputs["loss"]