Model interpretation

Hello, thanks for this great package. 
I am trying to reproduce the findings of your paper but I couldn't see the interpretation part in the repo, could you point me to this or provide some code on how to carry out the interpretation? 
Thanks!