Tidzo's picture
End of training
70ad200
raw
history blame
209 Bytes
{
"epoch": 6.72,
"total_flos": 1.2388649195611423e+18,
"train_loss": 1.7003454405163962,
"train_runtime": 3667.1764,
"train_samples_per_second": 4.535,
"train_steps_per_second": 0.034
}