all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": 1.212050437927246,
    "eval_logits/rejected": 1.967947244644165,
    "eval_logps/chosen": -469.93450927734375,
    "eval_logps/rejected": -550.9584350585938,
    "eval_loss": 0.5057631134986877,
    "eval_rewards/accuracies": 0.7350000143051147,
    "eval_rewards/chosen": -2.0144448280334473,
    "eval_rewards/margins": 1.0093281269073486,
    "eval_rewards/rejected": -3.023772954940796,
    "eval_runtime": 1180.5777,
    "eval_samples": 2000,
    "eval_samples_per_second": 1.694,
    "eval_steps_per_second": 0.847,
    "train_loss": 0.0026284047499827543,
    "train_runtime": 361.3652,
    "train_samples": 61135,
    "train_samples_per_second": 169.178,
    "train_steps_per_second": 21.145
}