all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.5725526809692383,
    "eval_logits/rejected": -2.5801451206207275,
    "eval_logps/chosen": -351.0971374511719,
    "eval_logps/rejected": -367.6675720214844,
    "eval_loss": 0.6707386374473572,
    "eval_rewards/accuracies": 0.5983333587646484,
    "eval_rewards/chosen": -0.28601789474487305,
    "eval_rewards/margins": 0.0687330961227417,
    "eval_rewards/rejected": -0.35475102066993713,
    "eval_runtime": 1581.2259,
    "eval_samples": 2994,
    "eval_samples_per_second": 1.893,
    "eval_steps_per_second": 0.237,
    "train_loss": 0.67484232365751,
    "train_runtime": 40604.4701,
    "train_samples": 19996,
    "train_samples_per_second": 0.492,
    "train_steps_per_second": 0.031
}