polejowska's picture
End of training
de607fd
raw
history blame
208 Bytes
{
"epoch": 2.97,
"total_flos": 2.8249710873624576e+16,
"train_loss": 1.9398403582365618,
"train_runtime": 40.5476,
"train_samples_per_second": 27.967,
"train_steps_per_second": 1.702
}