tangken333's picture
End of training
6f6439f verified
raw
history blame contribute delete
222 Bytes
{
"epoch": 2.987241531016278,
"total_flos": 4.6211312738788966e+17,
"train_loss": 0.7107904120832944,
"train_runtime": 62456.134,
"train_samples_per_second": 0.437,
"train_steps_per_second": 0.009
}