info.json

{
  "name": "Enterprise Strategic Tokenizer",
  "version": "3.0.0",
  "vocab_size": 96000,
  "training_time_minutes": 8.01,
  "training_data_gb": 2.08,
  "curated_lines": 6303705,
  "dropped_duplicates": 76458,
  "architecture": "NFKC + ByteLevel BPE",
  "created_at": "2026-04-05 09:29:25"
}