tokenizer_config.json

{
  "tokenizer_class": "MorphemeTokenizer",
  "model_type": "splade",
  "vocab_size": 57030,
  "max_length": 128,
  "do_lower_case": false,
  "seg_temp_train": 1.0,
  "seg_temp_eval": 0.0,
  "min_freq_word": 2,
  "min_freq_morph": 3
}