all_results.json

{
    "epoch": 0.9880609304240429,
    "eval_logits/chosen": 9.238293647766113,
    "eval_logits/rejected": 10.049151420593262,
    "eval_logps/chosen": -114.94315338134766,
    "eval_logps/ref_chosen": -106.16099548339844,
    "eval_logps/ref_rejected": -119.78473663330078,
    "eval_logps/rejected": -129.38494873046875,
    "eval_loss": 0.6894497275352478,
    "eval_rewards/accuracies": 0.5833333134651184,
    "eval_rewards/chosen": -0.08782169967889786,
    "eval_rewards/margins": 0.008180392906069756,
    "eval_rewards/rejected": -0.09600208699703217,
    "eval_runtime": 152.328,
    "eval_samples": 1080,
    "eval_samples_per_second": 7.09,
    "eval_steps_per_second": 1.772,
    "total_flos": 0.0,
    "train_loss": 0.6920125889778137,
    "train_runtime": 8059.7946,
    "train_samples": 9715,
    "train_samples_per_second": 1.205,
    "train_steps_per_second": 0.009
}