Update the tokenizer which inserted added_tokens to vocab

fix(tokenizer): Also updates tokenizer.json to prevent any mismatch.

chore(root): Initial files upload.