all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.235046148300171,
    "eval_logits/rejected": -2.230177640914917,
    "eval_logps/chosen": -34.02177047729492,
    "eval_logps/rejected": -37.52760696411133,
    "eval_loss": 0.4478302299976349,
    "eval_rewards/accuracies": 0.5224252939224243,
    "eval_rewards/chosen": 0.010223164223134518,
    "eval_rewards/margins": 0.01901474967598915,
    "eval_rewards/rejected": -0.008791583590209484,
    "eval_runtime": 145.7484,
    "eval_samples": 343,
    "eval_samples_per_second": 2.353,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.3756082092012678,
    "train_runtime": 3251.4223,
    "train_samples": 3079,
    "train_samples_per_second": 0.947,
    "train_steps_per_second": 0.118
}