Delete tokenizer.json

#4

Removing stale tokenizer.json (32K vocab) as discussed in transformers PR #48973 so AutoTokenizer relies on vocab.txt (100K vocab) via tokenizer_config.json.

Modeling the Variation in Arabic org

Thanks @rabeya100X !
I believe this indeed is enough to fix the bug.

AMR-KELEG changed pull request status to merged

Sign up or log in to comment