{ "epoch": 1.0, "eval_log_odds_chosen": 1.1819437742233276, "eval_log_odds_ratio": -0.47540709376335144, "eval_logits/chosen": -0.8723087906837463, "eval_logits/rejected": -0.7670671343803406, "eval_logps/chosen": -0.7900527119636536, "eval_logps/rejected": -1.678329348564148, "eval_loss": 1.2826783657073975, "eval_nll_loss": 1.2283066511154175, "eval_rewards/accuracies": 0.7319767475128174, "eval_rewards/chosen": -0.07900191843509674, "eval_rewards/margins": 0.08882816880941391, "eval_rewards/rejected": -0.16783475875854492, "eval_runtime": 715.0059, "eval_samples": 54991, "eval_samples_per_second": 76.97, "eval_steps_per_second": 1.203 }