all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.238292694091797,
    "eval_logits/rejected": -2.2334160804748535,
    "eval_logps/chosen": -34.009239196777344,
    "eval_logps/rejected": -37.498390197753906,
    "eval_loss": 0.3797042965888977,
    "eval_rewards/accuracies": 0.5336378812789917,
    "eval_rewards/chosen": 0.02277994342148304,
    "eval_rewards/margins": 0.006376237142831087,
    "eval_rewards/rejected": 0.01640370488166809,
    "eval_runtime": 145.6289,
    "eval_samples": 343,
    "eval_samples_per_second": 2.355,
    "eval_steps_per_second": 0.295,
    "train_loss": 0.31703355428460356,
    "train_runtime": 3251.5033,
    "train_samples": 3079,
    "train_samples_per_second": 0.947,
    "train_steps_per_second": 0.118
}