mansee's picture
End of training
bdba499
raw
history blame
210 Bytes
{
"epoch": 9.99,
"total_flos": 1.3106102295963156e+19,
"train_loss": 0.3501510629954847,
"train_runtime": 8665.4247,
"train_samples_per_second": 60.918,
"train_steps_per_second": 0.475
}