samitizerxu's picture
End of training
d6023eb
raw
history blame contribute delete
208 Bytes
{
"epoch": 3.0,
"total_flos": 1.1433292766387896e+18,
"train_loss": 1.1250548786587184,
"train_runtime": 757.8766,
"train_samples_per_second": 60.687,
"train_steps_per_second": 0.475
}