{ "epoch": 1.0, "eval_logits/chosen": -2.821291208267212, "eval_logits/rejected": -2.818777084350586, "eval_logps/chosen": -31.391199111938477, "eval_logps/rejected": -34.970741271972656, "eval_loss": 0.8988423347473145, "eval_rewards/accuracies": 0.5772424936294556, "eval_rewards/chosen": -0.07612106204032898, "eval_rewards/margins": 0.11468725651502609, "eval_rewards/rejected": -0.19080831110477448, "eval_runtime": 112.7622, "eval_samples": 343, "eval_samples_per_second": 3.042, "eval_steps_per_second": 0.381 }