chchen's picture
End of training
70ebf5b verified
raw
history blame
223 Bytes
{
"epoch": 4.938271604938271,
"total_flos": 1.8334743535787213e+17,
"train_loss": 0.033509917281568054,
"train_runtime": 5004.178,
"train_samples_per_second": 0.405,
"train_steps_per_second": 0.025
}