ALM-AHME's picture
End of training
7ce3d03
{
"epoch": 6.94,
"total_flos": 7.375405476885369e+18,
"train_loss": 0.5960829522569425,
"train_runtime": 11023.3494,
"train_samples_per_second": 3.801,
"train_steps_per_second": 0.059
}