all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -1.0600295066833496,
    "eval_logits/rejected": -0.8026158809661865,
    "eval_logps/chosen": -368.7221374511719,
    "eval_logps/rejected": -858.111328125,
    "eval_loss": 0.4027799963951111,
    "eval_rewards/accuracies": 0.734375,
    "eval_rewards/chosen": -1.3388758897781372,
    "eval_rewards/margins": 2.8251354694366455,
    "eval_rewards/rejected": -4.164011001586914,
    "eval_runtime": 60.6412,
    "eval_samples": 505,
    "eval_samples_per_second": 8.328,
    "eval_steps_per_second": 0.264,
    "train_loss": 0.26888106432225967,
    "train_runtime": 2670.405,
    "train_samples": 10364,
    "train_samples_per_second": 3.881,
    "train_steps_per_second": 0.061
}