Strange contrastive loss function

I know that loss function does not seem true but can we infer something related to what the wrong point is? Or it is simply because of something wrong in my code. It is a positive part of the contrastive loss. Thanks a lot!
Screenshot 2020-05-09 at 12.54.12