polejowska's picture
End of training
f306bcb
raw
history blame
209 Bytes
{
"epoch": 1.98,
"total_flos": 2.1241704283294925e+17,
"train_loss": 0.7265663851391185,
"train_runtime": 142.6799,
"train_samples_per_second": 60.555,
"train_steps_per_second": 0.463
}