{ "epoch": 1.0, "eval_logits/chosen": -2.2319846153259277, "eval_logits/rejected": -2.227167844772339, "eval_logps/chosen": -33.76113510131836, "eval_logps/rejected": -37.29436492919922, "eval_loss": 0.49125906825065613, "eval_rewards/accuracies": 0.545265793800354, "eval_rewards/chosen": 0.19139321148395538, "eval_rewards/margins": 0.03581271693110466, "eval_rewards/rejected": 0.15558050572872162, "eval_runtime": 145.6689, "eval_samples": 343, "eval_samples_per_second": 2.355, "eval_steps_per_second": 0.295 }