Progress

#1
by Gabriel - opened

Hej hur går det med denna modellen är den klar för att fine-tunea med?

National Library of Sweden / KBLab org

Hi,
no this model is (as far as I know) just a one-off test.
If you want to have a BERT-like model that has been trained like RoBERTa with much more data for much longer, with similar hyperparameters, then use one of our megatron-bert models.

Sign up or log in to comment