chchen's picture
End of training
14d9730 verified
raw
history blame contribute delete
222 Bytes
{
"epoch": 4.938271604938271,
"total_flos": 1.8817065711747072e+17,
"train_loss": 0.03361496642231941,
"train_runtime": 5228.875,
"train_samples_per_second": 0.387,
"train_steps_per_second": 0.024
}