d1_math_fasttext / all_results.json
sedrickkeh's picture
End of training
5c12a7f verified
{
"epoch": 5.0,
"total_flos": 5.015512062164468e+18,
"train_loss": 0.34359321679904875,
"train_runtime": 58237.925,
"train_samples_per_second": 2.713,
"train_steps_per_second": 0.021
}