{ "epoch": 1.0, "eval_logits/chosen": 97.7850112915039, "eval_logits/rejected": 97.75206756591797, "eval_logps/chosen": -33.28022384643555, "eval_logps/rejected": -37.269412994384766, "eval_loss": 2457.578857421875, "eval_rewards/accuracies": 0.5892857313156128, "eval_rewards/chosen": -0.008370478637516499, "eval_rewards/margins": 0.004658386576920748, "eval_rewards/rejected": -0.013028863817453384, "eval_runtime": 103.6308, "eval_samples": 343, "eval_samples_per_second": 3.31, "eval_steps_per_second": 0.415 }