Prachetas's picture
End of training
c5621a3
{
"epoch": 14.4,
"total_flos": 2.7800491620335616e+17,
"train_loss": 0.5843003829320271,
"train_runtime": 517.5663,
"train_samples_per_second": 22.49,
"train_steps_per_second": 0.174
}