ALM-AHME's picture
End of training
99536f2
{
"epoch": 12.0,
"total_flos": 2.0923685902916665e+19,
"train_loss": 0.1486589150943784,
"train_runtime": 11149.9067,
"train_samples_per_second": 6.848,
"train_steps_per_second": 0.214
}