all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.227961301803589,
    "eval_logits/rejected": -2.2231268882751465,
    "eval_logps/chosen": -34.07651138305664,
    "eval_logps/rejected": -37.578250885009766,
    "eval_loss": 0.6956681609153748,
    "eval_rewards/accuracies": 0.5157807469367981,
    "eval_rewards/chosen": -0.029370848089456558,
    "eval_rewards/margins": 0.013771643862128258,
    "eval_rewards/rejected": -0.043142493814229965,
    "eval_runtime": 145.5641,
    "eval_samples": 343,
    "eval_samples_per_second": 2.356,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.6291831790626823,
    "train_runtime": 3251.1508,
    "train_samples": 3079,
    "train_samples_per_second": 0.947,
    "train_steps_per_second": 0.118
}