ALM-AHME's picture
End of training
b462145
{
"epoch": 9.98,
"total_flos": 8.390377142257582e+18,
"train_loss": 0.465066796070651,
"train_runtime": 8763.5167,
"train_samples_per_second": 4.166,
"train_steps_per_second": 0.13
}