openthoughts3_100k_qwen25_1b_bsz256_lr2e5_epochs5/all_results.json

{
    "epoch": 5.0,
    "total_flos": 7122204608430080.0,
    "train_loss": 1.082116018475779,
    "train_runtime": 36273.766,
    "train_samples_per_second": 13.784,
    "train_steps_per_second": 0.054
}