a1_code_stackexchange / all_results.json
sedrickkeh's picture
End of training
6a131bf verified
{
"epoch": 4.982278481012658,
"total_flos": 1.1129855183851356e+18,
"train_loss": 0.46361072231235545,
"train_runtime": 41853.6475,
"train_samples_per_second": 3.775,
"train_steps_per_second": 0.029
}