all_results.json

{
    "epoch": 0.998691442030882,
    "eval_dpo_lambda": 1.0,
    "eval_logits/chosen": -1.0166884660720825,
    "eval_logits/rejected": -1.052775502204895,
    "eval_logps/chosen": -344.9288330078125,
    "eval_logps/rejected": -368.1219177246094,
    "eval_loss": 0.5514757037162781,
    "eval_rewards/accuracies": 0.7329999804496765,
    "eval_rewards/chosen": -0.44281935691833496,
    "eval_rewards/margins": 0.5280917286872864,
    "eval_rewards/rejected": -0.9709112048149109,
    "eval_runtime": 719.0315,
    "eval_samples": 2000,
    "eval_samples_per_second": 2.782,
    "eval_steps_per_second": 0.695,
    "total_flos": 0.0,
    "train_loss": 0.5871124714300567,
    "train_runtime": 50801.4362,
    "train_samples": 61134,
    "train_samples_per_second": 1.203,
    "train_steps_per_second": 0.009
}