all_results.json

{
    "epoch": 2.999297541394882,
    "eval_logits/chosen": 0.004835990257561207,
    "eval_logits/rejected": 0.06824379414319992,
    "eval_logps/chosen": -1.2670738697052002,
    "eval_logps/rejected": -1.4042190313339233,
    "eval_loss": 1.6698354482650757,
    "eval_rewards/accuracies": 0.5474777221679688,
    "eval_rewards/chosen": -1.2670738697052002,
    "eval_rewards/margins": 0.13714538514614105,
    "eval_rewards/rejected": -1.4042190313339233,
    "eval_runtime": 33.6516,
    "eval_samples": 1345,
    "eval_samples_per_second": 39.968,
    "eval_semantic_entropy": 0.8052493929862976,
    "eval_steps_per_second": 10.014,
    "total_flos": 0.0,
    "train_loss": 1.640899122824931,
    "train_runtime": 28500.0022,
    "train_samples": 59790,
    "train_samples_per_second": 6.294,
    "train_steps_per_second": 0.197
}