all_results.json

{
    "epoch": 0.9221902017291066,
    "eval_logits/chosen": -0.010529891587793827,
    "eval_logits/rejected": 0.1464616358280182,
    "eval_logps/chosen": -128.34133911132812,
    "eval_logps/ref_chosen": -139.74691772460938,
    "eval_logps/ref_rejected": -176.36155700683594,
    "eval_logps/rejected": -164.8606719970703,
    "eval_loss": 0.4340026378631592,
    "eval_rewards/accuracies": 0.43888887763023376,
    "eval_rewards/chosen": 0.1140558049082756,
    "eval_rewards/margins": -0.0009530979441478848,
    "eval_rewards/rejected": 0.11500890552997589,
    "eval_runtime": 144.7424,
    "eval_samples": 1080,
    "eval_samples_per_second": 7.462,
    "eval_sft_loss": 0.43400266766548157,
    "eval_steps_per_second": 1.865,
    "total_flos": 0.0,
    "train_loss": 0.45325408279895785,
    "train_runtime": 617.7917,
    "train_samples": 1387,
    "train_samples_per_second": 2.245,
    "train_steps_per_second": 0.016
}