all_results.json

{
    "epoch": 0.9221902017291066,
    "eval_logits/chosen": -0.04302170127630234,
    "eval_logits/rejected": 0.1195269301533699,
    "eval_logps/chosen": -138.40455627441406,
    "eval_logps/ref_chosen": -138.20062255859375,
    "eval_logps/ref_rejected": -174.97462463378906,
    "eval_logps/rejected": -175.22572326660156,
    "eval_loss": 0.6929154992103577,
    "eval_rewards/accuracies": 0.5379629731178284,
    "eval_rewards/chosen": -0.0020393093582242727,
    "eval_rewards/margins": 0.00047187993186526,
    "eval_rewards/rejected": -0.002511189319193363,
    "eval_runtime": 144.5655,
    "eval_samples": 1080,
    "eval_samples_per_second": 7.471,
    "eval_steps_per_second": 1.868,
    "total_flos": 0.0,
    "train_loss": 0.6933929324150085,
    "train_runtime": 650.6499,
    "train_samples": 1387,
    "train_samples_per_second": 2.132,
    "train_steps_per_second": 0.015
}