all_results.json

{
    "epoch": 2.992914501653283,
    "eval_dpo_loss": 0.7098360657691956,
    "eval_logits": -0.5715497136116028,
    "eval_logps": -81.13369750976562,
    "eval_loss": 578.1592407226562,
    "eval_objective": 566.3954467773438,
    "eval_ranking_simple": 0.5362318754196167,
    "eval_regularize": 0.5898738503456116,
    "eval_runtime": 367.9228,
    "eval_samples": 5790,
    "eval_samples_per_second": 15.737,
    "eval_steps_per_second": 1.313,
    "total_flos": 0.0,
    "train_loss": 356.3672220056707,
    "train_runtime": 34691.2758,
    "train_samples": 50802,
    "train_samples_per_second": 4.393,
    "train_steps_per_second": 0.03
}