{ "epoch": 1.0, "eval_logits/chosen": -2.2434651851654053, "eval_logits/rejected": -2.2385833263397217, "eval_logps/chosen": -34.03584671020508, "eval_logps/rejected": -37.496219635009766, "eval_loss": 0.7054064273834229, "eval_rewards/accuracies": 0.472591370344162, "eval_rewards/chosen": -0.001035186112858355, "eval_rewards/margins": -0.017356980592012405, "eval_rewards/rejected": 0.01632179319858551, "eval_runtime": 145.8229, "eval_samples": 343, "eval_samples_per_second": 2.352, "eval_steps_per_second": 0.295 }