all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -4.456838607788086,
    "eval_logits/rejected": -4.443554401397705,
    "eval_logps/chosen": -375.13671875,
    "eval_logps/rejected": -394.703125,
    "eval_loss": 0.6773372888565063,
    "eval_rewards/accuracies": 0.5686666369438171,
    "eval_rewards/chosen": -1.5442434549331665,
    "eval_rewards/margins": 0.13949742913246155,
    "eval_rewards/rejected": -1.6837408542633057,
    "eval_runtime": 1623.421,
    "eval_samples": 2994,
    "eval_samples_per_second": 1.844,
    "eval_steps_per_second": 0.231,
    "train_loss": 0.6475976420174225,
    "train_runtime": 42677.4758,
    "train_samples": 19996,
    "train_samples_per_second": 0.469,
    "train_steps_per_second": 0.029
}