all_results.json

{
    "epoch": 0.9993060374739764,
    "eval_logits/chosen": 1.1335889101028442,
    "eval_logits/rejected": 3.4289302825927734,
    "eval_logps/chosen": -786.9862670898438,
    "eval_logps/rejected": -1287.4072265625,
    "eval_loss": 0.23935413360595703,
    "eval_rewards/accuracies": 0.8825187683105469,
    "eval_rewards/chosen": -3.9893157482147217,
    "eval_rewards/margins": 4.835687637329102,
    "eval_rewards/rejected": -8.825004577636719,
    "eval_runtime": 388.0309,
    "eval_samples": 8491,
    "eval_samples_per_second": 21.882,
    "eval_steps_per_second": 0.343,
    "total_flos": 0.0,
    "train_loss": 0.32248947852187687,
    "train_runtime": 20768.7862,
    "train_samples": 184443,
    "train_samples_per_second": 8.881,
    "train_steps_per_second": 0.035
}