openr1 / all_results.json
sedrickkeh's picture
End of training
a77fc1e verified
{
"epoch": 4.990689013035381,
"total_flos": 8.620546826447094e+18,
"train_loss": 0.380745083004681,
"train_runtime": 51129.1279,
"train_samples_per_second": 3.36,
"train_steps_per_second": 0.007
}