tokenizer_config.json

{
  "tokenizer_class": "PreTrainedTokenizerFast",
  "vocab_size": 62720
}