So, did this experiment turn out useful? Does it work well with the new native VSA in the Model Sparse Attention node?
· Sign up or log in to comment
