{
  "id": "refinedweb",
  "name": "RefinedWeb",
  "identifiers": {
    "huggingface": [
      "tiiuae/falcon-refinedweb"
    ]
  },
  "builder": "Technology Innovation Institute (TII)",
  "release_date": {
    "value": "2023-06-01",
    "source": "https://arxiv.org/abs/2306.01116",
    "status": "partial",
    "note": "arXiv v1 date of the paper."
  },
  "availability": {
    "value": "partial",
    "checked": "2026-09-24",
    "source": "https://huggingface.co/datasets/tiiuae/falcon-refinedweb",
    "note": "Paper: 'We publicly release an extract of 600 billion tokens' of the five-trillion-token dataset."
  },
  "content": {
    "value": "Filtered, deduplicated English web text from CommonCrawl, ~5 trillion tokens.",
    "source": "https://arxiv.org/abs/2306.01116",
    "status": "recorded"
  },
  "primary_sources": [
    "https://arxiv.org/abs/2306.01116"
  ],
  "record_history": [
    {
      "date": "2026-09-24",
      "change": "created from primary sources (dataset records ruling)",
      "by": "wilson-pruitt + claude"
    }
  ]
}