all_results.json

{
    "epoch": 0.9880609304240429,
    "eval_logits/chosen": 9.191950798034668,
    "eval_logits/rejected": 9.997398376464844,
    "eval_logps/chosen": -115.93839263916016,
    "eval_logps/ref_chosen": -106.16099548339844,
    "eval_logps/ref_rejected": -119.78473663330078,
    "eval_logps/rejected": -130.08287048339844,
    "eval_loss": 0.6909545063972473,
    "eval_rewards/accuracies": 0.539814829826355,
    "eval_rewards/chosen": -0.09777405112981796,
    "eval_rewards/margins": 0.0052073062397539616,
    "eval_rewards/rejected": -0.10298136621713638,
    "eval_runtime": 171.06,
    "eval_samples": 1080,
    "eval_samples_per_second": 6.314,
    "eval_steps_per_second": 1.578,
    "total_flos": 0.0,
    "train_loss": 0.6924255990982056,
    "train_runtime": 7990.9765,
    "train_samples": 9715,
    "train_samples_per_second": 1.216,
    "train_steps_per_second": 0.009
}