qwen3-1.7b-dabstep-reasoning-108-fixed-reasoning-sharegpt-sft/train_results.json

{
    "epoch": 5.0,
    "total_flos": 852618117120.0,
    "train_loss": 0.9887651102883476,
    "train_runtime": 133.1517,
    "train_samples_per_second": 3.793,
    "train_steps_per_second": 0.263
}