{ "epoch": 1.0, "eval_logits/chosen": 0.7713278532028198, "eval_logits/rejected": 1.3411734104156494, "eval_logps/chosen": -549.3906860351562, "eval_logps/rejected": -626.2533569335938, "eval_loss": 0.487976998090744, "eval_rewards/accuracies": 0.726190447807312, "eval_rewards/chosen": -2.8615000247955322, "eval_rewards/margins": 1.0698496103286743, "eval_rewards/rejected": -3.931349515914917, "eval_runtime": 475.1451, "eval_samples": 2000, "eval_samples_per_second": 4.209, "eval_steps_per_second": 0.177 }