GRANTHE2761's picture
End of training
b37ddc1
{
"epoch": 3.0,
"total_flos": 1.8124066505760768e+18,
"train_loss": 0.3941639829100224,
"train_runtime": 487.9007,
"train_samples_per_second": 149.416,
"train_steps_per_second": 0.584
}