all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.2231976985931396,
    "eval_logits/rejected": -2.218374013900757,
    "eval_logps/chosen": -34.05440139770508,
    "eval_logps/rejected": -37.5644416809082,
    "eval_loss": 0.9781046509742737,
    "eval_rewards/accuracies": 0.5170266032218933,
    "eval_rewards/chosen": -0.01588033325970173,
    "eval_rewards/margins": 0.02237490564584732,
    "eval_rewards/rejected": -0.0382552333176136,
    "eval_runtime": 145.5341,
    "eval_samples": 343,
    "eval_samples_per_second": 2.357,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.8298485310046704,
    "train_runtime": 3249.7236,
    "train_samples": 3079,
    "train_samples_per_second": 0.947,
    "train_steps_per_second": 0.118
}