tokenizer/tokenizer_metadata.json

{
  "vocab_size": 200000,
  "training_time_minutes": 144.34882674217224,
  "timestamp": 1763844808.7371492,
  "missing_tokens": []
}