all_results.json

{
    "epoch": 0.9221902017291066,
    "eval_logits/chosen": -0.019660519436001778,
    "eval_logits/rejected": 0.1411634236574173,
    "eval_logps/chosen": -139.75112915039062,
    "eval_logps/ref_chosen": -140.15744018554688,
    "eval_logps/ref_rejected": -176.90542602539062,
    "eval_logps/rejected": -176.54791259765625,
    "eval_loss": 0.6929076910018921,
    "eval_rewards/accuracies": 0.5249999761581421,
    "eval_rewards/chosen": 0.004063152242451906,
    "eval_rewards/margins": 0.0004878933250438422,
    "eval_rewards/rejected": 0.003575259121134877,
    "eval_runtime": 143.2325,
    "eval_samples": 1080,
    "eval_samples_per_second": 7.54,
    "eval_steps_per_second": 1.885,
    "total_flos": 0.0,
    "train_loss": 0.6933379590511322,
    "train_runtime": 621.366,
    "train_samples": 1387,
    "train_samples_per_second": 2.232,
    "train_steps_per_second": 0.016
}