amjadfqs's picture
End of training
dc3bfc1
{
"epoch": 5.0,
"total_flos": 1.1595881296429056e+18,
"train_loss": 0.1904253252960266,
"train_runtime": 7888.6282,
"train_samples_per_second": 4.518,
"train_steps_per_second": 0.03
}