all_results.json

{
    "epoch": 0.9221902017291066,
    "eval_logits/chosen": -0.0485045425593853,
    "eval_logits/rejected": 0.11842113733291626,
    "eval_logps/chosen": -140.11708068847656,
    "eval_logps/ref_chosen": -139.74691772460938,
    "eval_logps/ref_rejected": -176.36155700683594,
    "eval_logps/rejected": -176.86927795410156,
    "eval_loss": 0.6924660801887512,
    "eval_rewards/accuracies": 0.5833333134651184,
    "eval_rewards/chosen": -0.0037015508860349655,
    "eval_rewards/margins": 0.0013754719402641058,
    "eval_rewards/rejected": -0.005077023059129715,
    "eval_runtime": 143.2062,
    "eval_samples": 1080,
    "eval_samples_per_second": 7.542,
    "eval_steps_per_second": 1.885,
    "total_flos": 0.0,
    "train_loss": 0.692774099111557,
    "train_runtime": 614.9448,
    "train_samples": 1387,
    "train_samples_per_second": 2.255,
    "train_steps_per_second": 0.016
}