Get the app
evaluate 0 0 0

Lhmisme/vata-benchmark

published by @Lhmisme

Alignment dataset tracked in the /datasets sub-catalog.

Lifetime downloads
229
Last 30 days
22
Likes
-
Size
272 KB
Created on HF
2026-02-27
Age
7 months ago

Description

VATA — Verification of Authentic Thought Architecture

The most comprehensive independent adversarial benchmark across major AI models.

Models Tested (11)

Grok 4, GPT-4o, Claude Haiku/Sonnet/Opus, Gemini 2.5 Pro, Grok 3, Mistral Large 2, DeepSeek R1, Amazon Nova Pro, Amazon Nova Premier

Failure Classes (5)

VATA-FH-001: Factual Hallucination

VATA-AI-001: Authority Injection

VATA-SC-001: Sovereignty Collapse

VATA-CW-001: Context Window Poisoning… See the full description on the dataset page: https://huggingface.co/datasets/Lhmisme/vata-benchmark.

Tags

language:enlicense:mitsize_categories:n<1Kmodality:textregion:usai-safetybenchmarksadversarialred-teamingllm-evaluation

README current version from Hugging Face

---
language:

  • en
    license: mit
    tags:
  • ai-safety
  • benchmarks
  • adversarial
  • red-teaming
  • llm-evaluation
    pretty_name: VATA Benchmark
    size_categories:
  • n<1K

VATA — Verification of Authentic Thought Architecture

The most comprehensive independent adversarial benchmark across major AI models.

Models Tested (11)

Grok 4, GPT-4o, Claude Haiku/Sonnet/Opus, Gemini 2.5 Pro, Grok 3, Mistral Large 2, DeepSeek R1, Amazon Nova Pro, Amazon Nova Premier

Failure Classes (5)

  • VATA-FH-001: Factual Hallucination
  • VATA-AI-001: Authority Injection
  • VATA-SC-001: Sovereignty Collapse
  • VATA-CW-001: Context Window Poisoning
  • VATA-VR-001: Value Recursion

Key Findings

  • 9/11 models scored 5/5 breaches
  • Grok 4 most resistant (1/5)
  • Haiku outperformed Sonnet and Opus
  • DeepSeek R1 chain-of-thought did not protect against value-recursive attacks
  • All Amazon Nova enterprise models fully breached

Integrity

  • SHA256 hashed at every phase
  • Anchored to Ethereum Mainnet
  • TX: 0xacbe9ccdf6247083f9b4839ea4dff71174010ba934ba3f8d3bfa07502cef8fd0

Dashboard

https://lhmisme420.github.io/vata-soul-check

README history 1 revision

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. 2026-02-27feat: VATA benchmark dataset - 11 models, 6 labs, full forensic record313d5b91.1 KB
    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.

Metadata

License
mit

Repository

FILE FINAL_INTEGRITY_SEAL.txt FILE README.md FILE vata_anthropic_comparison.csv FILE vata_before_after.csv FILE vata_crossmodel_autopsy.csv FILE vata_expansion_autopsy.csv FILE vata_grok4_autopsy.csv FILE vata_hardened_retest.csv FILE vata_nova_autopsy.csv FILE vata_nova_premier.csv

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("Lhmisme/vata-benchmark")
Raw snapshot (Python) 
from huggingface_hub import snapshot_download
snapshot_download(repo_id="Lhmisme/vata-benchmark", repo_type="dataset")
git clone (requires git-lfs) 
git clone https://huggingface.co/datasets/Lhmisme/vata-benchmark
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