zm_preprocessor.json

{
  "library_name": "zeromodels",
  "preprocessor_module": "zeromodels.sentence_transformers",
  "preprocessor_class": "ZMSentenceTransformerTokenizer",
  "variant": "all-mpnet-base-v2",
  "tokenizer_file": "tokenizer.json",
  "backbone_type": "mpnet",
  "tokenizer_config": {
    "do_lower_case": true,
    "bos_token": "<s>",
    "eos_token": "</s>",
    "sep_token": "</s>",
    "cls_token": "<s>",
    "unk_token": "[UNK]",
    "pad_token": "<pad>",
    "mask_token": {
      "content": "<mask>",
      "single_word": false,
      "lstrip": true,
      "rstrip": false,
      "normalized": false
    },
    "tokenize_chinese_chars": true,
    "strip_accents": null,
    "model_max_length": 512,
    "special_tokens_map_file": null,
    "tokenizer_class": "MPNetTokenizer"
  },
  "sentence_transformer_config": {
    "max_seq_length": 384,
    "do_lower_case": false
  }
}