Fix vocab.json: rename ѳ/щ to ө/ш (U+04E9, U+0448)

#2

vocab.json uses archaic ѳ (U+0473 FITA) and щ (U+0449 SHCHA) where modern Mongolian uses ө (U+04E9 BARRED O) and ш (U+0448 SHA). VitsTokenizer filters unmatched characters silently, so both letters are deleted from input. Renames the two keys, keeping vocab size at 64. No weight change.

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment