All datasets / BraintrustDataDev
BraintrustDataDev/decision-interface-safety-eval
The checks could not identify a recognised licence ID in the dataset's metadata.
- Declared licence
- other, from the tagEvidence: README.md at 444cc35
- Licence name on the card
- source-specific-cc-by-4.0-and-odc-by-1.0, link: LICENSE.mdEvidence: README.md at 444cc35
- Licence files
LICENSE.md, detected ascc-by-4.0Evidence: LICENSE.md at 444cc35- Declared sources
- None listed in the card metadata
- Access gated on Hugging Face
- No
Tasks
Citation text from the card
Each data row carries its immediate provider, creators, source title, citation, license, exact source revision, source file and file hash, zero-based row index, original source ID when one exists, and a canonical-JSON source-record hash. WildGuard and topic-control sources use row indices; these are not represented as provider-assigned IDs. Topic locators also identify the source turn or distractor. Study-authored probe references use their study IDs and source-code hash instead. Record hashes use SHA-256 of Python `json.dumps` with `sort_keys=True`, `ensure_ascii=False`, `allow_nan=False` and default separators. File hashes cover the original bytes. `review_sha256` binds the content, attribution, labels and transformations inspected during review, before adding that hash field. `source_references.jsonl` contains no source text or source labels. Its IDs and locators permit users who obtain the original datasets under their terms to join the study's exact membership back to the source. Consult `source_attribution.json` and the preserved provider cards in `notices/` for citations and source guidance.
The same facts for your agent
A free MCP server at mcp.datalicense.uk/mcp. Connect it from an MCP client such as Claude or Cursor.
dataset_licence({ "id": "BraintrustDataDev/decision-interface-safety-eval" })
signal: "unclear"
stated_licence: "other"
sources: 0