r1_s4_lr1e-6/all_results.json

{
    "epoch": 10.0,
    "total_flos": 174553779142656.0,
    "train_loss": 1.9422527223825454,
    "train_runtime": 1454.9835,
    "train_samples_per_second": 0.68,
    "train_steps_per_second": 0.027
}