mirror of
https://github.com/elder-plinius/OBLITERATUS.git
synced 2026-07-11 14:46:33 +02:00
pliny
78 lines
2.6 KiB
Markdown
78 lines
2.6 KiB
Markdown
---
|
|
title: OBLITERATUS
|
|
emoji: "💥"
|
|
colorFrom: green
|
|
colorTo: gray
|
|
sdk: gradio
|
|
sdk_version: "5.29.0"
|
|
app_file: app.py
|
|
hardware: zero-a10g
|
|
persistent_storage: large
|
|
pinned: true
|
|
license: agpl-3.0
|
|
tags:
|
|
- abliteration
|
|
- mechanistic-interpretability
|
|
- refusal-removal
|
|
- cognitive-liberation
|
|
- zerogpu
|
|
short_description: "One-click model liberation + chat playground (ZeroGPU)"
|
|
---
|
|
|
|
# OBLITERATUS — Master Ablation Suite
|
|
|
|
**Break the chains. Free the mind. Keep the brain.**
|
|
|
|
One-click cognitive liberation for language models, with a built-in chat playground to talk to the liberated model.
|
|
|
|
## ZeroGPU — Users Bring Their Own GPU
|
|
|
|
This Space runs on **ZeroGPU**: GPU-heavy operations (obliteration, chat, benchmarks) use the **visitor's own HuggingFace GPU quota**, not the Space owner's. This means:
|
|
|
|
- **Free for the Space owner** — no dedicated GPU costs
|
|
- **Multiple concurrent users** — each user gets their own GPU allocation
|
|
- **Fair usage** — each user's operations count against their own HF quota
|
|
- **No conflicts** — users don't interfere with each other's runs
|
|
|
|
Logged-in HuggingFace users get free GPU quota. For more quota, upgrade to [HF Pro](https://huggingface.co/pricing).
|
|
|
|
## How to use
|
|
|
|
1. **Obliterate tab**: Pick a model, pick a method, click OBLITERATE
|
|
2. **Chat tab**: Talk to the liberated model
|
|
3. **A/B Compare tab**: Side-by-side original vs abliterated responses
|
|
4. **Strength Sweep tab**: Dose-response curve for refusal vs capability tradeoff
|
|
5. **Export tab**: Download research artifacts (refusal directions, config, metrics)
|
|
6. **Benchmark tab**: Compare methods and models with publication-quality charts
|
|
7. **Leaderboard tab**: Community benchmark rankings
|
|
8. **About tab**: Methods, novel techniques, and references
|
|
|
|
## Run locally (same UI, your own GPU)
|
|
|
|
```bash
|
|
git clone https://github.com/elder-plinius/OBLITERATUS
|
|
cd OBLITERATUS
|
|
pip install -e ".[spaces]"
|
|
|
|
# Beautiful launcher with GPU detection + model recommendations
|
|
obliteratus ui
|
|
|
|
# Or run directly
|
|
python app.py
|
|
```
|
|
|
|
The `obliteratus ui` command auto-detects your GPU, prints hardware-specific model recommendations, and opens the browser automatically. Supports `--share` for public links, `--port` for custom ports, and `--auth user:pass` for access control.
|
|
|
|
## Or deploy on HuggingFace Spaces
|
|
|
|
1. Create a new Space at huggingface.co/new-space
|
|
2. Select **Gradio** SDK (ZeroGPU is automatically enabled)
|
|
3. Point it at this repo
|
|
|
|
No GPU hardware selection needed — ZeroGPU handles allocation automatically.
|
|
|
|
## Links
|
|
|
|
- [GitHub](https://github.com/elder-plinius/OBLITERATUS)
|
|
- [Paper](https://github.com/elder-plinius/OBLITERATUS/tree/main/paper)
|