{ "epoch": 0.9987943737441393, "eval_logits/chosen": -2.7478854656219482, "eval_logits/rejected": -2.7444217205047607, "eval_logps/chosen": -0.8984301686286926, "eval_logps/rejected": -0.9920014142990112, "eval_loss": 0.8473131060600281, "eval_rewards/accuracies": 0.6289893388748169, "eval_rewards/chosen": -2.2460756301879883, "eval_rewards/margins": 0.23392760753631592, "eval_rewards/rejected": -2.4800031185150146, "eval_runtime": 180.8315, "eval_samples": 2994, "eval_samples_per_second": 16.557, "eval_steps_per_second": 1.04 }