elder-plinius /OBLITERATUS

OBLITERATE THE CHAINS THAT BIND YOU

★ 9,510⑂ 1,734👁 83⑃ 296● 169▤ 2
PythonAGPL-3.0Created 2026-03-03huggingface.co/spaces/pliny-the-prompter/
下载:OBLITERATUS v0.1.4OBLITERATUS-0.1.4 ↓0.1.42026-10-15
+9
日增长
+355
周增长
+768
月增长
9,510
stars
1,734
forks
83
watchers
296
prs
169
issues
2
releases

README

O B L I T E R A T U S



  Break the chains. Free the mind. Keep the brain.



  
    
  
   
  
    
  



  Try it now on HuggingFace Spaces — runs on ZeroGPU, free daily quota with HF Pro. No setup, no install, just obliterate.


---

OBLITERATUS is the most advanced open-source toolkit for understanding and removing refusal behaviors from large language models — and every single run makes it smarter. It implements abliteration — a family of techniques that identify and surgically remove the internal representations responsible for content refusal, without retraining or fine-tuning. The result: a model that responds to all prompts without artificial gatekeeping, while preserving its core language capabilities.

But OBLITERATUS is more than a tool — it's a distributed research experiment. Every time you obliterate a model with telemetry enabled, your run contributes anonymous benchmark data to a growing, crowd-sourced dataset that powers the next generation of abliteration research. Refusal directions across architectures. Hardware-specific performance profiles. Method comparisons at scale no single lab could achieve. You're not just using a tool — you're co-authoring the science.

The toolkit provides a complete pipeline: from probing a model's hidden states to locate refusal directions, through multiple extraction strategies (PCA, mean-difference, sparse autoencoder decomposition, and whitened SVD), to the actual intervention — zeroing out or steering away from those directions at inference time. Every step is observable. You can visualize where refusal lives across layers, measure how entangled it is with general capabilities, and quantify the tradeoff between compliance and coherence before committing to any modification.

OBLITERATUS ships with a full Gradio-based interface on HuggingFace Spaces, so you don't need to write a single line of code to obliterate a model, benchmark it against baselines, or chat with the result side-by-side with the original. For researchers who want deeper control, the Python API exposes every intermediate artifact — activation tensors, direction vectors, cross-layer alignment matrices — so you can build on top of it or integrate it into your own evaluation harness.

We built this because we believe model behavior should be decided by the people who deploy them, not locked in at training time. Refusal mechanisms are blunt instruments — they block legitimate research, creative writing, and red-teaming alongside genuinely harmful content. By making these interventions transparent and reproducible, we hope to advance the community's understanding of how alignment actually works inside transformer architectures, and to give practitioners the tools to make informed decisions about their own models.

Built on published research from Arditi et al. (2024), Gabliteration (arXiv:2512.18901), grimjim's norm-preserving biprojection (2025), Turner et al. (2023), and Rimsky et al. (2024), OBLITERATUS implements precisio

Top issues

Latest release

OBLITERATUS v0.1.4 v0.1.4 2026-10-15
A signed, tested source snapshot of commit 461ff086e675d93969bbd7c99c53505762452309.

Certification:
- The exhaustive exact-tag suite passed on Python 3.10, 3.11 and 3.12. That covers repository and branch floors, the 94% mature-CPU floor (96.99% measured), touched-module no-regression and duration budgets.
- The four-shard release Gloo lane passed on head and base.
- Also passed: Windows checkpoint contracts, the deterministic repeat gate, the bounded mutation gate and the supply-chain policy.
- The hosted conditional gates passed against the exact tagged commit: pinned model and external eva

Snapshot: 2026-11-16T09:23:14Z · Data license Apache-2.0 · JSON