all_results.json

{
    "epoch": 0.9922480620155039,
    "eval_logits/chosen": -2.135279655456543,
    "eval_logits/rejected": -2.0949225425720215,
    "eval_logps/chosen": -164.1279296875,
    "eval_logps/rejected": -183.2969970703125,
    "eval_loss": 0.6877921223640442,
    "eval_rewards/accuracies": 0.53125,
    "eval_rewards/chosen": -0.0802411139011383,
    "eval_rewards/margins": 0.007847035303711891,
    "eval_rewards/rejected": -0.08808814734220505,
    "eval_runtime": 2.3656,
    "eval_samples": 100,
    "eval_samples_per_second": 42.272,
    "eval_steps_per_second": 0.845,
    "total_flos": 0.0,
    "train_loss": 0.6876659728586674,
    "train_runtime": 1604.6168,
    "train_samples": 33000,
    "train_samples_per_second": 20.566,
    "train_steps_per_second": 0.04
}