all_results.json

{
    "epoch": 1.999247554552295,
    "eval_logits/chosen": -0.2070515751838684,
    "eval_logits/rejected": 0.09358108043670654,
    "eval_logps/chosen": -117.798583984375,
    "eval_logps/rejected": -908.2842407226562,
    "eval_loss": 0.04265851154923439,
    "eval_rewards/accuracies": 1.0,
    "eval_rewards/chosen": -0.013141422532498837,
    "eval_rewards/margins": 7.185807228088379,
    "eval_rewards/rejected": -7.198948383331299,
    "eval_runtime": 193.1837,
    "eval_samples": 546,
    "eval_samples_per_second": 2.826,
    "eval_steps_per_second": 1.413,
    "total_flos": 0.0,
    "train_loss": 0.01645123219139964,
    "train_runtime": 7014.6347,
    "train_samples": 2657,
    "train_samples_per_second": 0.758,
    "train_steps_per_second": 0.024
}