{ "epoch": 1.0, "eval_logits/chosen": 97.65992736816406, "eval_logits/rejected": 97.62660217285156, "eval_logps/chosen": -33.290000915527344, "eval_logps/rejected": -37.37261199951172, "eval_loss": 0.994398832321167, "eval_rewards/accuracies": 0.6004983186721802, "eval_rewards/chosen": -0.008468217216432095, "eval_rewards/margins": 0.005592645611613989, "eval_rewards/rejected": -0.014060864225029945, "eval_runtime": 103.8235, "eval_samples": 343, "eval_samples_per_second": 3.304, "eval_steps_per_second": 0.414 }