all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.340519428253174,
    "eval_logits/rejected": -2.282647132873535,
    "eval_logps/chosen": -357.6859436035156,
    "eval_logps/rejected": -441.0806884765625,
    "eval_loss": 0.5274128913879395,
    "eval_rewards/accuracies": 0.78515625,
    "eval_rewards/chosen": -0.9042894840240479,
    "eval_rewards/margins": 0.8080792427062988,
    "eval_rewards/rejected": -1.7123687267303467,
    "eval_runtime": 103.8477,
    "eval_samples": 2000,
    "eval_samples_per_second": 19.259,
    "eval_steps_per_second": 0.308,
    "total_flos": 0.0,
    "train_loss": 0.5690948443535047,
    "train_runtime": 12671.4295,
    "train_samples": 61134,
    "train_samples_per_second": 4.825,
    "train_steps_per_second": 0.151
}