all_results.json

{
    "epoch": 0.9880609304240429,
    "eval_logits/chosen": 9.246881484985352,
    "eval_logits/rejected": 10.068053245544434,
    "eval_logps/chosen": -74.92684936523438,
    "eval_logps/ref_chosen": -106.16099548339844,
    "eval_logps/ref_rejected": -119.78473663330078,
    "eval_logps/rejected": -87.84820556640625,
    "eval_loss": 0.26139113306999207,
    "eval_rewards/accuracies": 0.47962963581085205,
    "eval_rewards/chosen": 0.3123414218425751,
    "eval_rewards/margins": -0.007023785263299942,
    "eval_rewards/rejected": 0.3193652033805847,
    "eval_runtime": 165.9841,
    "eval_samples": 1080,
    "eval_samples_per_second": 6.507,
    "eval_sft_loss": 0.26139113306999207,
    "eval_steps_per_second": 1.627,
    "total_flos": 0.0,
    "train_loss": 0.29937881012757617,
    "train_runtime": 8120.0447,
    "train_samples": 9715,
    "train_samples_per_second": 1.196,
    "train_steps_per_second": 0.009
}