tokenizer.json

{
  "model_type": "bert",
  "do_lower_case": true,
  "strip_accents": false,
  "pad_token_id": 0,
  "wordpieces_prefix": "##"
}