moreover18's picture
End of training
95f16d1
raw
history blame contribute delete
209 Bytes
{
"epoch": 2.99,
"total_flos": 5.830187330643444e+18,
"train_loss": 0.18055563719094206,
"train_runtime": 6946.1332,
"train_samples_per_second": 10.85,
"train_steps_per_second": 0.169
}