{ "epoch": 0.9981298423724285, "eval_logits/chosen": 0.21227747201919556, "eval_logits/rejected": 0.22475622594356537, "eval_logps/chosen": -166.8560791015625, "eval_logps/rejected": -174.9294891357422, "eval_loss": 0.6347980499267578, "eval_rewards/accuracies": 0.5766128897666931, "eval_rewards/chosen": -0.7856369614601135, "eval_rewards/margins": 0.09609751403331757, "eval_rewards/rejected": -0.8817344307899475, "eval_runtime": 166.8441, "eval_samples_per_second": 11.753, "eval_steps_per_second": 0.186 }