Kinda sucks because we normally use British spelling, but it just looks weird and confusing otherwise... same with tokenizer and all other library internals. So this is sort of the "official policy" for now.