{ "epoch": 4.988190836088805, "eval_dpo_loss": 0.679865837097168, "eval_logits": -1.2945246696472168, "eval_logps": -88.41614532470703, "eval_loss": 187.94732666015625, "eval_objective": 183.8509979248047, "eval_ranking_idealized": 0.6024844646453857, "eval_ranking_idealized_expo": 0.5232919454574585, "eval_ranking_simple": 0.532608687877655, "eval_regularize": 0.416841596364975, "eval_runtime": 521.0952, "eval_samples": 5790, "eval_samples_per_second": 11.111, "eval_steps_per_second": 0.927, "eval_wo_beta": 15.995294570922852, "total_flos": 0.0, "train_loss": 67.88850653388283, "train_runtime": 74214.1269, "train_samples": 50802, "train_samples_per_second": 3.423, "train_steps_per_second": 0.024 }