dataset.json

{
  "dataset_name": "Auralis_packed_dataset",
  "token_format": "uint32",
  "tokenizer": "Qwen/Qwen2.5-1.5B",
  "existing_tokens_before_run": 176544387,
  "new_tokens_this_run": 4141,
  "total_tokens_after_run": 176548528,
  "documents_tokenized": 1,
  "duplicates_skipped": 0,
  "records_skipped": 0,
  "existing_shards_preserved": true,
  "new_shards_start": 16
}