all_results.json

{
    "epoch": 0.05592059275828324,
    "eval_logits/chosen": -1.2117185592651367,
    "eval_logits/rejected": -1.3683098554611206,
    "eval_logps/chosen": -1.3815547227859497,
    "eval_logps/rejected": -2.264155149459839,
    "eval_loss": 0.9751168489456177,
    "eval_rewards/accuracies": 0.7612612843513489,
    "eval_rewards/chosen": -3.4538867473602295,
    "eval_rewards/margins": 2.206501007080078,
    "eval_rewards/rejected": -5.66038703918457,
    "eval_runtime": 344.9163,
    "eval_samples": 1328,
    "eval_samples_per_second": 3.85,
    "eval_steps_per_second": 0.322,
    "total_flos": 0.0,
    "train_loss": 1.163502004146576,
    "train_runtime": 4144.9448,
    "train_samples": 42917,
    "train_samples_per_second": 0.579,
    "train_steps_per_second": 0.048
}