{ "epoch": 1.0, "eval_logits/chosen": -2.819918155670166, "eval_logits/rejected": -2.7931315898895264, "eval_logps/chosen": -319.5545654296875, "eval_logps/rejected": -329.6432189941406, "eval_loss": 0.19092892110347748, "eval_pred_label": 19507.943359375, "eval_rewards/accuracies": 0.6940000057220459, "eval_rewards/chosen": -3.757920265197754, "eval_rewards/margins": 2.9776601791381836, "eval_rewards/rejected": -6.735579967498779, "eval_runtime": 855.9781, "eval_samples": 2000, "eval_samples_per_second": 2.337, "eval_steps_per_second": 0.292, "eval_use_label": 12554.0556640625 }