shivarama23's picture
End of training
d4de32a
{
"epoch": 10.0,
"total_flos": 2.23703900872704e+16,
"train_loss": 0.4094994068145752,
"train_runtime": 71.4678,
"train_samples_per_second": 12.593,
"train_steps_per_second": 0.14
}