chchen's picture
End of training
d61202e verified
raw
history blame contribute delete
224 Bytes
{
"epoch": 4.994228549442093,
"total_flos": 8.169046237996646e+17,
"train_loss": 0.035229922796976974,
"train_runtime": 38657.4974,
"train_samples_per_second": 1.345,
"train_steps_per_second": 0.084
}