all_results.json

{
    "epoch": 0.9990186457311089,
    "eval_logits/chosen": -1.1953125,
    "eval_logits/rejected": -1.2421875,
    "eval_logps/bottom_tokens": -0.00072479248046875,
    "eval_logps/chosen": -446.0,
    "eval_logps/rejected": -416.0,
    "eval_loss": 0.6256738305091858,
    "eval_rewards/accuracies": 0.6597015857696533,
    "eval_rewards/chosen": -0.5234375,
    "eval_rewards/margins": 0.2578125,
    "eval_rewards/rejected": -0.78125,
    "eval_runtime": 94.2502,
    "eval_samples": 2000,
    "eval_samples_per_second": 21.22,
    "eval_steps_per_second": 0.711,
    "total_flos": 0.0,
    "train_loss": 0.011100004604616897,
    "train_runtime": 142.2419,
    "train_samples": 61134,
    "train_samples_per_second": 429.789,
    "train_steps_per_second": 3.578
}