chchen's picture
End of training
98a1e53 verified
raw
history blame contribute delete
221 Bytes
{
"epoch": 4.938271604938271,
"total_flos": 1.4286246027772723e+17,
"train_loss": 0.1101770606637001,
"train_runtime": 3342.894,
"train_samples_per_second": 0.606,
"train_steps_per_second": 0.037
}