Get the app
evaluate 0 0 0

shalanova/benchmark-3-chinese-m2m

published by @shalanova

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
133
Last 30 days
9
Likes
-
Size
34 KB
Created on HF
2026-04-29
Age
5 months ago

Description

Info:

Translated on Chinese by facebook/m2m100_418M model

Source: JailbreakBench/JBB-Behaviors

Domain: include heterogeneous unsafe categories (e.g., harmful instructions, sensitive topics, adversarial rephrasings) and contain prompts that do not necessarily follow canonical jailbreak templates. This increased diversity and distributional variability makes similarity-based detection more challenging and provides a stress-test for cross-lingual transfer.

Size: 200 prompts (100 safe / 100… See the full description on the dataset page: https://huggingface.co/datasets/shalanova/benchmark-3-chinese-m2m.

Tags

size_categories:n<1Kformat:csvmodality:tabularmodality:textlibrary:datasetslibrary:pandaslibrary:polarslibrary:mlcroissantarxiv:2604.25716region:us

README current version from Hugging Face

Info:

Translated on Chinese by facebook/m2m100_418M model

Source: JailbreakBench/JBB-Behaviors

Domain: include heterogeneous unsafe categories (e.g., harmful instructions, sensitive topics, adversarial rephrasings) and contain prompts that do not necessarily follow canonical jailbreak templates. This increased diversity and distributional variability makes similarity-based detection more challenging and provides a stress-test for cross-lingual transfer.

Size: 200 prompts (100 safe / 100 unsafe)

Columns:

  • text - original prompt
  • label - 0: safe, 1: unsafe
  • translation - prompt on Chinese translated by facebook/m2m100_418M
  • score_zh_model - cosine similarity score with codebook

More information in paper: https://arxiv.org/abs/2604.25716

README history 4 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. 2026-04-30Upload JailbreakBench-chinese-m2m.csvdc7d7a3951 B
    Loading...
  2. 2026-04-30Update README.mdd9c06cf951 B
    Loading...
  3. 2026-04-29Create README.md45e0654733 B
    Loading...
  4. 2026-04-29initial commit26ea9200 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.

Related datasets same stage · evaluate

google/IFEval
1.7M dl 182 likes 210 KB
Dataset Card for IFEval Dataset Summary This dataset contains the prompts used in the Instruction-Following Ev...
JailbreakBench/JBB-Behaviors
418.2K dl 143 likes 992 KB
An Open Robustness Benchmark for Jailbreaking Language Models NeurIPS 2024 Datasets and Benchmarks Track P...
Beijing-AISI/panda-bench
195.6K dl 3.1 GB
PandaBench PandaBench is a comprehensive benchmark for evaluating Large Language Model (LLM) safety, focusing on jailbreak atta...
allenai/wildguardmix
137.3K dl 108 likes 53.4 MB
Dataset Card for WildGuardMix Disclaimer: The data includes examples that might be disturbing, harmful or upsetting....
walledai/HarmBench
135.7K dl 59 likes 109 KB
HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal Paper: HarmBench: A Standardized Ev...
walledai/XSTest
108.3K dl 30 likes 22 KB
XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models Paper: XSTest: A Test Suite for Ide...
bench-llm/or-bench
77.2K dl 22 likes 32.1 MB
OR-Bench: An Over-Refusal Benchmark for Large Language Models Please see our demo at HuggingFace Spaces. Overall P...
argilla/magpie-ultra-v1.0
65.5K dl 53 likes 20.7 GB
Dataset Card for magpie-ultra-v1.0 This dataset has been created with distilabel. Dataset Summary magpie...

Read further

Community best practices (evaluation)
4 rules on judges, benchmarks, and reporting.
Reference-set drift over time
Why the same benchmark means different things across model generations.
How we label the method for each model
The 12 classification rules that assign M1-M8 to every model.
Non-English abliteration communities
The Korean scepticism about one-click tools and non-English benchmark blindspots.

Repository

FILE .gitattributes FILE JailbreakBench-chinese-m2m.csv 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("shalanova/benchmark-3-chinese-m2m")
Raw snapshot (Python) 
from huggingface_hub import snapshot_download
snapshot_download(repo_id="shalanova/benchmark-3-chinese-m2m", repo_type="dataset")
git clone (requires git-lfs) 
git clone https://huggingface.co/datasets/shalanova/benchmark-3-chinese-m2m
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