{ "epoch": 0.01, "eval_logits/chosen": -2.410409688949585, "eval_logits/rejected": -2.363970994949341, "eval_logps/chosen": -276.4896240234375, "eval_logps/rejected": -257.1393127441406, "eval_loss": 0.6865259408950806, "eval_rewards/accuracies": 0.593500018119812, "eval_rewards/chosen": 0.03312591835856438, "eval_rewards/margins": 0.014345898292958736, "eval_rewards/rejected": 0.018780020996928215, "eval_runtime": 615.2299, "eval_samples": 2000, "eval_samples_per_second": 3.251, "eval_steps_per_second": 0.406 }