chchen's picture
End of training
874210e verified
{
"epoch": 4.994228549442093,
"total_flos": 8.169046237996646e+17,
"train_loss": 0.03533345094547594,
"train_runtime": 44857.0674,
"train_samples_per_second": 1.159,
"train_steps_per_second": 0.072
}