{ "epoch": 1.0, "eval_logits/chosen": 0.9867910742759705, "eval_logits/rejected": 1.0411394834518433, "eval_logps/chosen": -110.39482879638672, "eval_logps/rejected": -161.07052612304688, "eval_loss": 0.617142915725708, "eval_pred_label": 0.0, "eval_rewards/accuracies": 0.37109375, "eval_rewards/chosen": -0.4648258090019226, "eval_rewards/margins": 0.3739502727985382, "eval_rewards/rejected": -0.8387760519981384, "eval_runtime": 125.4187, "eval_samples": 2000, "eval_samples_per_second": 15.947, "eval_steps_per_second": 0.255, "eval_use_label": 0.0 }