polejowska's picture
End of training
d3dcda9
raw
history blame
210 Bytes
{
"epoch": 9.98,
"total_flos": 1.0716320475504968e+18,
"train_loss": 0.24182906033414783,
"train_runtime": 827.7733,
"train_samples_per_second": 52.188,
"train_steps_per_second": 0.399
}