all_results.json

{
    "epoch": 0.9995419147961521,
    "eval_logits/chosen": -0.6407761573791504,
    "eval_logits/rejected": -0.6567662358283997,
    "eval_logps/chosen": -200.93621826171875,
    "eval_logps/rejected": -276.8379821777344,
    "eval_loss": 0.30047258734703064,
    "eval_rewards/accuracies": 0.8699421882629395,
    "eval_rewards/chosen": 1.6737332344055176,
    "eval_rewards/margins": 2.166951894760132,
    "eval_rewards/rejected": -0.4932188391685486,
    "eval_runtime": 252.7931,
    "eval_samples": 2763,
    "eval_samples_per_second": 10.93,
    "eval_steps_per_second": 1.369,
    "total_flos": 0.0,
    "train_loss": 0.37405444834616947,
    "train_runtime": 8738.9435,
    "train_samples": 34924,
    "train_samples_per_second": 3.996,
    "train_steps_per_second": 0.125
}