cosmo5b-04/_aggregate.json

{
  "engine": "tokforge",
  "prefix": "cosmo5b-04",
  "corpus": "cosmo",
  "version": "5b",
  "shards": 3,
  "docs": 376288,
  "tokens": 271321339,
  "bytes": 542642678,
  "files": [
    "data-00000-of-00003.npy",
    "data-00001-of-00003.npy",
    "data-00002-of-00003.npy"
  ],
  "source_range": [
    91,
    92
  ],
  "source_dataset": "HuggingFaceTB/smollm-corpus",
  "source_config": "cosmopedia-v2",
  "created_at": "2026-08-18T13:47:41Z"
}