tokenizer_config.json

{
  "backend": "tokenizers",
  "clean_up_tokenization_spaces": false,
  "cls_token": "[CLS]",
  "is_local": false,
  "mask_token": "[MASK]",
  "merges_file": "artifacts/modernbert_bpe_tokenizer_artifacts/merges.txt",
  "model_max_length": 8192,
  "pad_token": "[PAD]",
  "sep_token": "[SEP]",
  "tokenizer_class": "PreTrainedTokenizerFast",
  "unk_token": "[UNK]",
  "model_input_names": [
    "input_ids",
    "attention_mask"
  ]
}