tokenizer.json

{
  "type": "BertTokenizer",
  "vocab_file": "vocab.txt",
  "lower_case": true
}