all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.243932008743286,
    "eval_logits/rejected": -2.239053249359131,
    "eval_logps/chosen": -34.03658676147461,
    "eval_logps/rejected": -37.49867630004883,
    "eval_loss": 1.0139278173446655,
    "eval_rewards/accuracies": 0.4663621187210083,
    "eval_rewards/chosen": -0.0014253630070015788,
    "eval_rewards/margins": -0.013985722325742245,
    "eval_rewards/rejected": 0.01256035827100277,
    "eval_runtime": 145.6648,
    "eval_samples": 343,
    "eval_samples_per_second": 2.355,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.9909477320584383,
    "train_runtime": 3254.3492,
    "train_samples": 3079,
    "train_samples_per_second": 0.946,
    "train_steps_per_second": 0.118
}