{ "epoch": 0.9988221436984688, "eval_logits/chosen": 0.19542913138866425, "eval_logits/rejected": 0.3584606647491455, "eval_logps/chosen": -420.2536926269531, "eval_logps/ref_chosen": -287.60125732421875, "eval_logps/ref_rejected": -265.3289489746094, "eval_logps/rejected": -453.9424133300781, "eval_loss": 0.5453774333000183, "eval_margin_dpo/margin_mean": 55.96099090576172, "eval_margin_dpo/margin_std": 56.519386291503906, "eval_model_preparation_time": 0.0013, "eval_runtime": 46.2276, "eval_samples": 2000, "eval_samples_per_second": 43.264, "eval_steps_per_second": 5.408 }