ALM-AHME's picture
End of training
7e246b4
{
"epoch": 6.94,
"total_flos": 7.375405476885369e+18,
"train_loss": 0.7435173462795955,
"train_runtime": 8012.7314,
"train_samples_per_second": 5.229,
"train_steps_per_second": 0.081
}