ALM-AHME's picture
End of training
d023f1f
raw
history blame
211 Bytes
{
"epoch": 11.98,
"total_flos": 1.9127019697307763e+19,
"train_loss": 0.1604562966202188,
"train_runtime": 12187.6968,
"train_samples_per_second": 8.861,
"train_steps_per_second": 0.277
}