all_results.json

{
    "epoch": 5.0,
    "eval_logits/chosen": -2.0124804973602295,
    "eval_logits/rejected": -1.8513933420181274,
    "eval_logps/chosen": -289.3787841796875,
    "eval_logps/rejected": -276.745849609375,
    "eval_loss": 0.059522345662117004,
    "eval_rewards/accuracies": 0.40299999713897705,
    "eval_rewards/chosen": -0.11340584605932236,
    "eval_rewards/margins": -0.004375707823783159,
    "eval_rewards/rejected": -0.10903014987707138,
    "eval_runtime": 704.211,
    "eval_samples": 2000,
    "eval_samples_per_second": 2.84,
    "eval_steps_per_second": 1.42,
    "train_loss": 0.3429992333650589,
    "train_runtime": 12841.5285,
    "train_samples": 61135,
    "train_samples_per_second": 0.389,
    "train_steps_per_second": 0.097
}