all_results.json

{
    "epoch": 0.9999183606825047,
    "eval_logits/chosen": -1.4790972471237183,
    "eval_logits/rejected": -1.4601374864578247,
    "eval_logps/chosen": -844.0220947265625,
    "eval_logps/rejected": -928.1929321289062,
    "eval_loss": 0.4941699206829071,
    "eval_rewards/accuracies": 0.7620000243186951,
    "eval_rewards/chosen": -3.3367679119110107,
    "eval_rewards/margins": 0.9227678179740906,
    "eval_rewards/rejected": -4.259535789489746,
    "eval_runtime": 300.0346,
    "eval_samples": 4000,
    "eval_samples_per_second": 13.332,
    "eval_steps_per_second": 0.833,
    "total_flos": 0.0,
    "train_loss": 0.5524635882372952,
    "train_runtime": 40329.8241,
    "train_samples": 195977,
    "train_samples_per_second": 4.859,
    "train_steps_per_second": 0.038
}