{
  "tokenizer": {
    "vocab_size": 8000,
    "min_frequency": 2,
    "special_tokens": [
      "<s>",
      "<pad>",
      "</s>",
      "<unk>",
      "<mask>"
    ],
    "dataset": "facebook/xnli/vi",
    "corpus_lines": 800404
  },
  "metrics": {
    "tokens_per_char": 0.2967440807085845,
    "fertility": 1.314215987511743,
    "avg_seq_len": 24.807666666666666,
    "vocab_coverage": 1.0,
    "total_sentences": 15000,
    "unique_word_types": 9556
  }
}