aup-fullft-kto_w1-w1lam9.68e-4-seed3/train_results.json

{
    "epoch": 3.0,
    "total_flos": 9.960196480656998e+17,
    "train_loss": 0.586677935045206,
    "train_runtime": 11045.1759,
    "train_samples_per_second": 13.419,
    "train_steps_per_second": 0.21
}