In Japanese, some hiragana and katakana characters can have diacritics to change their pronunciation - either ゙ (dots) or ゚ (circle). For example, ヒ (hi) becomes ビ (bi) or ピ (pi). However, with the Japanese variants of CJK characters added in 24w06a, all the diacritics resemble the circle. This makes characters that can have either diacritic look the same (such as ビ and ピ), as well as putting the circle diacritic on characters that can only use the dots (such as デ (de) ). See the attached screenshots for more examples.
Can confirm. Most likely caused by current design of kana in Unifont JP.