{ "epoch": 1.0, "eval_logits/chosen": 98.72113800048828, "eval_logits/rejected": 98.69998931884766, "eval_logps/chosen": -32.468719482421875, "eval_logps/rejected": -36.04634094238281, "eval_loss": 1.107681393623352, "eval_rewards/accuracies": 0.4987541437149048, "eval_rewards/chosen": -0.022987432777881622, "eval_rewards/margins": 0.04884451627731323, "eval_rewards/rejected": -0.07183194905519485, "eval_runtime": 103.7689, "eval_samples": 343, "eval_samples_per_second": 3.305, "eval_steps_per_second": 0.414 }