I managed to remove most of the styling information that slipped through training by zeroing out all but four layers. This model is exactly equivalent to using the regular v1 model with the following layer weights: 1,0,0,0,0,0,0,0,1,1,1,0,0,0,0,0,0

It works great for photorealistic models and styles that aren't like the original art. You can also use it to turn Fei into a girl or whatever -- I won't judge.