ALM-AHME's picture
End of training
41d6b1a
{
"epoch": 14.0,
"total_flos": 1.5805887922325717e+19,
"train_loss": 0.17515284487896893,
"train_runtime": 12010.6398,
"train_samples_per_second": 7.418,
"train_steps_per_second": 0.232
}