{ "epoch": 1.0, "eval_logits/chosen": 0.03850164636969566, "eval_logits/rejected": 0.8719873428344727, "eval_logps/chosen": -394.6778259277344, "eval_logps/rejected": -466.36431884765625, "eval_loss": 0.5035951733589172, "eval_rewards/accuracies": 0.7698412537574768, "eval_rewards/chosen": -1.1057459115982056, "eval_rewards/margins": 0.9401550889015198, "eval_rewards/rejected": -2.04590106010437, "eval_runtime": 207.7886, "eval_samples": 2000, "eval_samples_per_second": 9.625, "eval_steps_per_second": 0.303 }