{ "epoch": 0.995910949568378, "eval_logits/chosen": -13.09176254272461, "eval_logits/rejected": -12.961532592773438, "eval_logps/chosen": -1.1284456253051758, "eval_logps/rejected": -1.3546873331069946, "eval_loss": 3.8762307167053223, "eval_rewards/accuracies": 0.7110655903816223, "eval_rewards/chosen": -11.284457206726074, "eval_rewards/margins": 2.2624149322509766, "eval_rewards/rejected": -13.54687213897705, "eval_runtime": 86.4298, "eval_samples": 1941, "eval_samples_per_second": 22.458, "eval_steps_per_second": 1.412 }