all_results.json

{
    "epoch": 3.5474728389230044,
    "eval_dpo_loss": 0.7511767148971558,
    "eval_logits": -0.47565215826034546,
    "eval_logps": -82.51917266845703,
    "eval_loss": 0.5233561396598816,
    "eval_objective": 0.5224636197090149,
    "eval_ranking_idealized": 0.6030020713806152,
    "eval_ranking_idealized_expo": 0.5222567319869995,
    "eval_ranking_simple": 0.5253623127937317,
    "eval_regularize": 0.5224636197090149,
    "eval_runtime": 308.14,
    "eval_samples": 5790,
    "eval_samples_per_second": 18.79,
    "eval_steps_per_second": 1.567,
    "eval_wo_beta": 14.005528450012207,
    "total_flos": 0.0,
    "train_loss": 0.02363297004699707,
    "train_runtime": 6884.3268,
    "train_samples": 50802,
    "train_samples_per_second": 36.897,
    "train_steps_per_second": 0.256
}