lixiqi's picture
End of training
06109df
{
"epoch": 3.0,
"total_flos": 6.668732964123095e+18,
"train_loss": 1.15604233954634,
"train_runtime": 2196.8246,
"train_samples_per_second": 39.205,
"train_steps_per_second": 0.306
}