{ "epoch": 0.999630314232902, "eval_logits/chosen": -2.554856538772583, "eval_logits/rejected": -2.423921585083008, "eval_logps/chosen": -432.22442626953125, "eval_logps/rejected": -477.8772888183594, "eval_loss": 0.5821850299835205, "eval_rewards/accuracies": 0.6875, "eval_rewards/chosen": -1.6933226585388184, "eval_rewards/margins": 0.3464769423007965, "eval_rewards/rejected": -2.039799690246582, "eval_runtime": 322.3528, "eval_samples": 1976, "eval_samples_per_second": 6.13, "eval_steps_per_second": 0.385 }