keerthan2's picture
End of training
0e6649c
{
"epoch": 3.0,
"total_flos": 1.8124066505760768e+18,
"train_loss": 0.33303696230838176,
"train_runtime": 1123.47,
"train_samples_per_second": 64.888,
"train_steps_per_second": 0.507
}