Hi! I have a text classification task and I fine tune a DistilBERT for it. After executing trainer.train(), I saw a message that starts with TrainOutput. It contains some information such as train loss, but the the value of train loss in TrainOutput is different from the value of train loss for last epoch. My question is what TrainOutputis and why the value of train loss in it, is different from the value of train loss for last epoch. I searched about it but I didn’t find any information about it.
I appreciate any help 
