all_results.json

{
    "epoch": 4.988190836088805,
    "eval_dpo_loss": 0.6852555871009827,
    "eval_logits": -1.7323659658432007,
    "eval_logps": -99.90007019042969,
    "eval_loss": 0.4175891578197479,
    "eval_objective": 0.4183734059333801,
    "eval_ranking_idealized": 0.6570248007774353,
    "eval_ranking_idealized_expo": 0.5113636255264282,
    "eval_ranking_simple": 0.5237603187561035,
    "eval_regularize": 0.4183734059333801,
    "eval_runtime": 259.2163,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.337,
    "eval_steps_per_second": 0.934,
    "total_flos": 0.0,
    "train_loss": 0.19899855256080629,
    "train_runtime": 35117.4941,
    "train_samples": 50802,
    "train_samples_per_second": 7.233,
    "train_steps_per_second": 0.025
}