I had a look at the lists (hosted on Github, sourced from another post here). On the surface, the lists for China, HK, and TW are longer. But if we consider that these lists contain both English filters and their counterparts in Asian languages, by design they will be much longer.

For example, US is listed as 170 filters. You can see the majority of them in the HK list + their translations in Cantonese + specific terms for HK.

In China, where at least 5 "languages" are covered, including traditional Chinese, simplified Chinese, Japanese, rhymes, and pinyin, the 1,045 (6 times more than US' 170) does not seem excessive at all.

(Edit: corrected filter counts)