ALM-AHME's picture
End of training
6f9c3d3
{
"epoch": 4.96,
"total_flos": 8.138660625246413e+18,
"train_loss": 0.23672226468722027,
"train_runtime": 4894.1951,
"train_samples_per_second": 6.114,
"train_steps_per_second": 0.095
}