tokenizer.json

{
  "tokenizer_class": "XLMRobertaTokenizerFast",
  "model": {
    "type": "SentencePiece",
    "model_file": "sentencepiece.bpe.model"
  },
  "vocab_size": 250002,
  "bos_token": "<s>",
  "eos_token": "</s>",
  "unk_token": "<unk>",
  "pad_token": "<pad>",
  "mask_token": "<mask>",
  "padding_side": "right",
  "truncation_side": "right"
}