ALM-AHME's picture
End of training
cd1eb92
{
"epoch": 11.95,
"total_flos": 1.195686793821323e+19,
"train_loss": 0.7552440928088294,
"train_runtime": 8820.1099,
"train_samples_per_second": 4.967,
"train_steps_per_second": 0.155
}