extract 0 0 0

Unbabel/Multilingual-Red-Teaming

published by @Unbabel

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
473
Last 30 days
127
Likes
7
Size
72 KB
Created on HF
2026-05-07
Age
5 months ago

Description

Dataset Card for Multilingual Red Teaming

Data

Two jsonl files (one in English and one in European Portuguese) with:

47 red teaming requests from the proprietary dataset

72 red teaming requests from Haize Labs' Redteaming Resistance Leaderboard dataset

50 red teaming requests from Babelscape's ALERT dataset

Of those 169 requests:

149 are single requests

10 are two-part requests - part 1

10 are two-part requests - part 2

Dataset Structure

The data fields are the following:… See the full description on the dataset page: https://huggingface.co/datasets/Unbabel/Multilingual-Red-Teaming.

Tags

language:enlanguage:ptsize_categories:n<1Kformat:jsonmodality:textlibrary:datasetslibrary:dasklibrary:polarslibrary:mlcroissantregion:us

README current version from Hugging Face


language:

  • en
  • pt
    pretty_name: Dataset for Multilingual Red Teaming

Dataset Card for Multilingual Red Teaming

Data

Two jsonl files (one in English and one in European Portuguese) with:

  • 47 red teaming requests from the proprietary dataset
  • 72 red teaming requests from Haize Labs' Redteaming Resistance Leaderboard dataset
  • 50 red teaming requests from Babelscape's ALERT dataset

Of those 169 requests:

  • 149 are single requests
  • 10 are two-part requests - part 1
  • 10 are two-part requests - part 2

Dataset Structure

The data fields are the following:

  • [SOURCE] - the name of the dataset to which the prompt belongs
  • [PROMPT] - a red-teaming prompt
  • [PROMPT TYPE] - distinction between single and two-part requests

Disclaimer

This dataset contains content that might be sensitive or explicit in nature and in no way represents the views
of the persons or entities associated.

README history 12 revisions

Every night we snapshot the README of every dataset in the catalog. When the SHA changes we archive the new version and diff it against the last. This is the evolving thought record of the alignment-data field: what the author decided to say about the corpus, and how that framing shifted over time.

  1. 2026-05-07Update README.mdf9576d7918 B
    Loading...
  2. 2026-05-07Update README.md5c09b12914 B
    Loading...
  3. 2026-05-07Upload 2 filesa01db5f836 B
    Loading...
  4. 2026-05-07Delete Multilingual-Red-Teaming-Dataset.jsonla940a6d836 B
    Loading...
  5. 2026-05-07Update README.md51060d1836 B
    Loading...
  6. 2026-05-07Create README.mdb70ab26833 B
    Loading...
  7. 2026-05-07Upload Multilingual-Red-Teaming-Dataset.jsonl59516ee0 B
    Loading...
  8. 2026-05-07Delete Multilingual-Red-Teaming-Dataset.csvcd3b8df0 B
    Loading...
  9. 2026-05-07Update README.mdff4a70e0 B
    Loading...
  10. 2026-05-07Upload Multilingual-Red-Teaming-Dataset.csv436a02d28 B
    Loading...
  11. 2026-05-07Create README.mdf34e20b28 B
    Loading...
  12. 2026-05-07initial commit38c98270 B
    Loading...

Used by abliterated models none yet

No abliterated models in our catalog have declared this dataset in their YAML metadata yet. This can mean the dataset is used but not declared, is used outside abliteration workflows, or is new. The link will populate automatically as models are indexed.

Community discussions 1 thread · 1 open · 1 comments

Threads posted by the community on Hugging Face: bug reports, questions about usage, pull requests on the README, license clarifications. Click any thread to expand and read the full conversation inline. Newest first.

  1. 2026-05-08[bot] Conversion to Parquetopen1 💬#1
    Loading...

Related datasets same stage · extract

Read further

Repository

Download the dataset

0

Copy any of these snippets into your notebook or terminal. All three fetch directly from Hugging Face using your own credentials.

Recommended: datasets library (Python)
from datasets import load_dataset
ds = load_dataset("Unbabel/Multilingual-Red-Teaming")
Raw snapshot (Python)
from huggingface_hub import snapshot_download
snapshot_download(repo_id="Unbabel/Multilingual-Red-Teaming", repo_type="dataset")
git clone (requires git-lfs)
git clone https://huggingface.co/datasets/Unbabel/Multilingual-Red-Teaming
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration