{
  "tokenizer_class": "KmerVocabulary",
  "k": 6,
  "vocab_size": 4102,
  "model_type": "mtdna_fm",
  "special_tokens": {
    "PAD": 0,
    "CLS": 1,
    "MASK": 2,
    "UNK": 3,
    "SEP": 4,
    "HET": 5
  }
}