{ "epoch": 2.9974597798475866, "eval_logits/chosen": -14.433335304260254, "eval_logits/rejected": -14.40054702758789, "eval_logps/chosen": -1.4238022565841675, "eval_logps/rejected": -1.5122665166854858, "eval_loss": 1.4967381954193115, "eval_odds_ratio_loss": 0.7293583154678345, "eval_rewards/accuracies": 0.5038095116615295, "eval_rewards/chosen": -0.1423802226781845, "eval_rewards/margins": 0.008846436627209187, "eval_rewards/rejected": -0.15122665464878082, "eval_runtime": 447.4835, "eval_samples_per_second": 2.346, "eval_sft_loss": 1.4238022565841675, "eval_steps_per_second": 1.173 }