{ "epoch": 2.992914501653283, "eval_dpo_loss": 0.6784854531288147, "eval_logits": -1.4974991083145142, "eval_logps": -92.26131439208984, "eval_loss": 179.2621307373047, "eval_objective": 175.9752197265625, "eval_ranking_idealized": 0.6024844646453857, "eval_ranking_idealized_expo": 0.5232919454574585, "eval_ranking_simple": 0.5279502868652344, "eval_regularize": 0.3991592228412628, "eval_runtime": 368.4144, "eval_samples": 5790, "eval_samples_per_second": 15.716, "eval_steps_per_second": 1.311, "eval_wo_beta": 16.585630416870117, "total_flos": 0.0, "train_loss": 87.93023242011215, "train_runtime": 38512.4809, "train_samples": 50802, "train_samples_per_second": 3.957, "train_steps_per_second": 0.027 }