tanlq's picture
End of training
b180bca
{
"epoch": 3.0,
"total_flos": 1.1623391924884734e+19,
"train_loss": 0.39782794105462216,
"train_runtime": 35181.3105,
"train_samples_per_second": 4.264,
"train_steps_per_second": 0.133
}