Hi everyone,
I just used AutoNLP for the first time tor fine-tune a model on a question-answering tasks. The training worked without frictions, but I am looking for more information on how to read the loss function-results I get for the 15 models trained (see picture below). The lower the loss function the better, but is 2.6 a good value? Can anyone provide me a link to a more detailed documentation?
Thanks!
Chris