10k_MT5_small_sum-de_GNAD / train_results.json
Einmalumdiewelt's picture
End of training
dca79fe
{
"epoch": 10.0,
"train_loss": 2.648989501953125,
"train_runtime": 4314.4712,
"train_samples": 7000,
"train_samples_per_second": 16.224,
"train_steps_per_second": 1.622
}