diff --git a/swift/trainers/rlhf_trainer/grpo_trainer.py b/swift/trainers/rlhf_trainer/grpo_trainer.py index c6b74e5e4..666ddc3c8 100644 --- a/swift/trainers/rlhf_trainer/grpo_trainer.py +++ b/swift/trainers/rlhf_trainer/grpo_trainer.py @@ -521,6 +521,6 @@ def _get_per_token_logps(self, model, inputs): def evaluation_loop(self, *args, **kwargs): metric_key_prefix = kwargs['metric_key_prefix'] output = super().evaluation_loop(*args, **kwargs) - metrics = {f'{metric_key_prefix}_{key}': sum(val) / len(val) for key, val in self._metrics.items()} + metrics = {f'{metric_key_prefix}_{key}': sum(val) / len(val) for key, val in self._metrics['eval'].items()} output.metrics.update(metrics) return output