all_results.json

{
    "epoch": 1.97,
    "eval_logits/chosen": 95.56348419189453,
    "eval_logits/rejected": 89.52069091796875,
    "eval_logps/chosen": -420.5301818847656,
    "eval_logps/rejected": -450.6969909667969,
    "eval_loss": 0.462095707654953,
    "eval_rewards/accuracies": 0.7291666865348816,
    "eval_rewards/chosen": -2.9011573791503906,
    "eval_rewards/margins": 1.6201168298721313,
    "eval_rewards/rejected": -4.521275043487549,
    "eval_runtime": 47.0999,
    "eval_samples": 750,
    "eval_samples_per_second": 15.924,
    "eval_steps_per_second": 0.51,
    "train_loss": 0.38778070876231563,
    "train_runtime": 1146.4695,
    "train_samples": 6750,
    "train_samples_per_second": 11.775,
    "train_steps_per_second": 0.091
}