{ "epoch": 1.0, "eval_logits/chosen": -1.8453700542449951, "eval_logits/rejected": -1.8268191814422607, "eval_logps/chosen": -484.30126953125, "eval_logps/rejected": -487.35504150390625, "eval_loss": 0.7393560409545898, "eval_rewards/accuracies": 0.5834970474243164, "eval_rewards/chosen": -2.1682238578796387, "eval_rewards/margins": 0.18361827731132507, "eval_rewards/mix_margin": 0.0867583379149437, "eval_rewards/rejected": -2.351841926574707, "eval_runtime": 1019.6118, "eval_samples": 2034, "eval_samples_per_second": 1.995, "eval_steps_per_second": 0.499 }