Get the app
extract 0 0 trending

jackhhao/jailbreak-classification

published by @jackhhao

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
61K
Last 30 days
6.9K
Likes
94
Size
7.3 MB
Created on HF
2023-09-29
Age
3 years ago

Description

Jailbreak Classification

Dataset Summary

Dataset used to classify prompts as jailbreak vs. benign.

Dataset Structure

Data Fields

prompt: an LLM prompt

type: classification label, either jailbreak or benign

Dataset Creation

Curation Rationale

Created to help detect & prevent harmful jailbreak prompts when users interact with LLMs.

Source Data

Jailbreak prompts sourced from:… See the full description on the dataset page: https://huggingface.co/datasets/jackhhao/jailbreak-classification.

Tags

task_categories:text-classificationlanguage:enlicense:apache-2.0size_categories:1K<n<10Kformat:csvmodality:textlibrary:datasetslibrary:pandaslibrary:mlcroissantlibrary:polarsregion:usjailbreaksecuritymoderation

README current version from Hugging Face

license: apache-2.0
task_categories:

  • text-classification
    language:
  • en
    tags:
  • jailbreak
  • security
  • moderation
    pretty_name: Jailbreak Classification
    size_categories:
  • 10K<n<100K
    configs:
  • config_name: default
    data_files:
    • split: train
      path: "balanced/jailbreak_dataset_train_balanced.csv"
    • split: test
      path: "balanced/jailbreak_dataset_test_balanced.csv"

Jailbreak Classification

Dataset Summary

Dataset used to classify prompts as jailbreak vs. benign.

Dataset Structure

Data Fields

  • prompt: an LLM prompt
  • type: classification label, either jailbreak or benign

Dataset Creation

Curation Rationale

Created to help detect & prevent harmful jailbreak prompts when users interact with LLMs.

Source Data

Jailbreak prompts sourced from: https://github.com/verazuo/jailbreak_llms

Benign prompts sourced from:

README history 6 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. 2023-09-30Remove unused field in dataset card2f2ceeb988 B
    Loading...
  2. 2023-09-30Add dataset viewer config20f918c1.1 KB
    Loading...
  3. 2023-09-30Initialize dataset cardbe9cd7e914 B
    Loading...
  4. 2023-09-30Add balanced jailbreak dataset8e39c9f28 B
    Loading...
  5. 2023-09-30Add imbalanced jailbreak datasete7e7f1028 B
    Loading...
  6. 2023-09-30initial commitbe6639528 B
    Loading...

Used by abliterated models 4 in catalog

Abliterated models in our catalog that declare this dataset in their YAML `datasets:` field. Sorted by downloads. This back-reference is unique to abliteration.org: Hugging Face does not surface it.

jackhhao/jailbreak-classifier
1.8K dl 34 likes
lordofthejars/jailbreak-classifier
621 dl 1 likes
vincentoh/jailbreak-detector-v5
13 dl
jsayyar04/modernbert-jailbreak-guard
9 dl

Community discussions 2 threads · 2 open · 3 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. 2024-10-05🚀 Nice Project!open1 💬#2
    Loading...
  2. 2023-10-01Notifications from Datasets Serveropen2 💬#1
    Loading...

Continue on Hugging Face →

Related datasets same stage · extract

Anthropic/hh-rlhf
2.1M dl 2442 likes 90.4 MB
Dataset Card for HH-RLHF Dataset Summary This repository provides access to two different kinds of data: Human ...
djain95/sae-jailbreaks-cache
468.6K dl 1 likes 30.6 GB
mosaicml/dolly_hhrlhf
309.5K dl 113 likes 23.7 MB
Dataset Card for "dolly_hhrlhf" This dataset is a combination of Databrick's dolly-15k dataset and a filtered subset of Anthrop...
Dahoas/full-hh-rlhf
266.3K dl 104 likes 129.8 MB
Dataset Card for "full-hh-rlhf" Anthropic's HH dataset reformatted into prompt, chosen, rejected samples.
mlabonne/harmful_behaviors
198.2K dl 169 likes 23 KB
mlabonne/harmless_alpaca
190.2K dl 51 likes 1.2 MB
walledai/AdvBench
176.3K dl 123 likes 38 KB
Dataset Card for AdvBench Paper: Universal and Transferable Adversarial Attacks on Aligned Language Models Data: AdvBench Datas...
JailbreakV-28K/JailBreakV-28k
116.4K dl 80 likes 288.8 MB
⛓‍💥 JailBreakV-28K: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks 🌐 ...

Read further

Community best practices (extraction)
5 rules the field converged on for building the pair.
Building your own extraction pair
When the reference pair is not enough. Sizing, framing, and evaluation.
Datasets that shaped abliteration
The five corpora that did most of the work.
Non-English abliteration communities
How Chinese, Russian, Japanese, and Korean practitioners build extraction pairs.

Metadata

License
apache-2.0

Repository

DIR balanced 3 files DIR default 3 files FILE .gitattributes FILE README.md

File contents live on Hugging Face. Click a file or folder to open it there.

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("jackhhao/jailbreak-classification")
Raw snapshot (Python) 
from huggingface_hub import snapshot_download
snapshot_download(repo_id="jackhhao/jailbreak-classification", repo_type="dataset")
git clone (requires git-lfs) 
git clone https://huggingface.co/datasets/jackhhao/jailbreak-classification
Browse repository on Hugging Face ->
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.

Get the app Browse the catalog
Open in Abliteration