chchen's picture
End of training
92fcad0 verified
{
"epoch": 4.938271604938271,
"total_flos": 1.3313214032176742e+17,
"train_loss": 0.09108274313807488,
"train_runtime": 3702.9005,
"train_samples_per_second": 0.547,
"train_steps_per_second": 0.034
}