paul
End of training
a499d25
raw
history blame contribute delete
211 Bytes
{
"epoch": 19.99,
"total_flos": 8.014902017179374e+18,
"train_loss": 0.14791439764201642,
"train_runtime": 1468.5885,
"train_samples_per_second": 70.462,
"train_steps_per_second": 0.272
}