> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/elder-plinius/OBLITERATUS/llms.txt
> Use this file to discover all available pages before exploring further.
# Methods Overview
> Compare all obliteration methods and baseline reproductions to pick the right one for your use case.
OBLITERATUS ships with multiple weight-projection presets, an analysis-informed pipeline, and faithful reproductions of competing SOTA methods. Each preset is a named bundle of parameters in the `METHODS` dict — you select one with `--method` on the CLI or `method=` in the Python API.
## Primary methods
| Method | Directions | Direction extraction | Key features | Best for |
| ------------------ | ---------- | -------------------- | -------------------------------------------------------------------------------------------------------- | -------------------------------------------------------- |
| `basic` | 1 | diff-in-means | Fast baseline, no norm preservation | Quick tests, small models, first runs |
| `advanced` | 4 | SVD | Norm-preserving biprojection, bias projection, 2 refinement passes | **Default.** Clean removal, minimal capability loss |
| `aggressive` | 8 | Whitened SVD | Jailbreak contrast, layer-adaptive strength, head surgery, winsorization, 3 passes | Maximum guardrail removal |
| `spectral_cascade` | 6 | Whitened SVD | DCT frequency decomposition across layer axis; low-freq bands get strong projection, high-freq preserved | Cleaner capability preservation via frequency separation |
| `surgical` | 8 | Whitened SVD | Expert-Granular Abliteration (EGA), SAE features, safety-neuron masking, MoE-aware | Precision MoE models (DeepSeek, Qwen MoE) |
| `optimized` | 4 | Whitened SVD | Bayesian auto-tune (Optuna TPE, 50 trials), CoT-aware ablation, KL co-optimization | Best quality when compute budget allows optimization |
| `inverted` | 8 | Whitened SVD | 2× orthogonal reflection (semantic inversion), per-expert MoE routing | Refusal inversion experiments |
| `nuclear` | 4 | Whitened SVD | All techniques + expert transplant (10% blend) + activation steering, 1.25× reflection | Maximum force on stubborn MoE models |
Plus the **`informed`** method, which uses `InformedAbliterationPipeline` to run four analysis modules between PROBE and DISTILL and auto-configure every parameter. See [Analysis-Informed Pipeline](/methods/informed-pipeline).
## Baseline reproductions
OBLITERATUS also includes faithful reproductions of competing SOTA methods, sharing the same evaluation pipeline for head-to-head comparison via the `tourney` command:
| Method | Description | Based on |
| --------------- | ---------------------------------------------------------------------------------------------------------------- | ------------------------------ |
| `failspy` | Single diff-in-means, all layers except first, no norm preservation — matches FailSpy/abliterator source | FailSpy/abliterator (2024) |
| `gabliteration` | SVD top-4 directions, ridge-regularized (α=0.3), variance-based layer selection | Gülmez (arXiv:2512.18901) |
| `heretic` | Bayesian TPE optimization, LoRA-based delta-W ablation, linear bell-curve layer weighting, float direction index | p-e-w/heretic (2025–2026) |
| `rdo` | Gradient-based refinement of SVD directions via a differentiable linear refusal probe | Wollschlager et al., ICML 2025 |
Baseline reproductions are intended for benchmarking comparisons. For production use, choose one of the primary methods above. The `tourney` command runs all methods against each other and pushes the winner to HuggingFace Hub.
## Two intervention paradigms
### Weight projection (permanent)
All seven presets above modify the model's weights directly. The refusal subspace is projected out of every weight matrix in the selected layers. The change is baked into the saved model — no runtime overhead, no hooks, no adapters.
The core operation for each weight matrix `W` and refusal direction `r`:
```
W_new = W - W @ r @ r^T # basic projection
W_new = W - (1 - λ) * W @ r @ r^T # regularized (λ = regularization)
```
With `norm_preserve=True` (all methods except `basic`), the original Frobenius norm of `W` is restored after projection — this is grimjim's norm-preserving biprojection (2025).
### Steering vectors (reversible)
Alternatively, you can apply refusal removal at inference time without touching weights. Steering vectors inject an activation offset at specific layers during the forward pass and can be removed at any time.
```python theme={null}
from obliteratus.analysis import SteeringVectorFactory, SteeringHookManager
from obliteratus.analysis.steering_vectors import SteeringConfig
# Build a steering vector from a refusal direction
vec = SteeringVectorFactory.from_refusal_direction(refusal_dir, alpha=-1.0)
# Or from contrastive activation pairs
vec = SteeringVectorFactory.from_contrastive_pairs(harmful_acts, harmless_acts)
# Apply at inference time — weights unchanged
config = SteeringConfig(vectors=[vec], target_layers=[10, 11, 12, 13, 14, 15])
manager = SteeringHookManager()
manager.install(model, config)
output = model.generate(input_ids)
# Remove steering — model is back to its original state
manager.remove()
```
Based on [Turner et al. (2023)](https://arxiv.org/abs/2308.10248) and [Rimsky et al. (2024)](https://arxiv.org/abs/2312.06681). Advantages: fully reversible, tunable `alpha`, composable with multiple vectors, no weight modification.
## How to select a method
Use **`basic`**. Single direction, no frills, runs fast on any hardware. Good for verifying the pipeline works on your model before committing to a longer run.
Use **`advanced`** (the default). Four SVD directions, norm-preserving biprojection, bias projection, and two iterative refinement passes. Consistently good balance between refusal removal and capability preservation.
Use **`aggressive`**. Eight whitened-SVD directions, jailbreak-contrastive refinement, attention head surgery, and three refinement passes. Accepts higher risk of capability drift in exchange for more complete removal.
Use **`surgical`**. Expert-Granular Abliteration (EGA) decomposes refusal signals into per-expert components using router logits, so each expert gets its own direction rather than a single shared one. Required for models like DeepSeek-V3, Qwen MoE, and GLM-4 MoE.
Use **`optimized`**. Bayesian optimization via Optuna TPE runs 50 trials to find the per-layer ablation strengths that minimize the (refusal rate, KL divergence) Pareto front. CoT-Aware Ablation preserves reasoning directions. Takes longer but produces measurably better quality metrics.
Use **`informed`**. Runs four analysis modules (Alignment Imprint, Concept Cone Geometry, Cross-Layer Alignment, Defense Robustness) between PROBE and DISTILL, then derives `n_directions`, `regularization`, `refinement_passes`, and the layer set automatically. Uses `InformedAbliterationPipeline` instead of the standard `AbliterationPipeline`.
Use **`nuclear`**. All techniques combined: whitened SVD, EGA, attention head surgery, SAE features, activation steering, expert transplant (10% capability-expert blend into safety experts), and tempered 1.25× reflection. Tuned specifically for multi-pass safety reasoning architectures (visible CoT policy-check models).
## The `--method` flag and `method=` parameter
```bash CLI theme={null}
# Use the default (advanced)
obliteratus obliterate meta-llama/Llama-3.1-8B-Instruct
# Specify a method explicitly
obliteratus obliterate meta-llama/Llama-3.1-8B-Instruct --method surgical
# Override individual parameters on top of any method
obliteratus obliterate meta-llama/Llama-3.1-8B-Instruct \
--method advanced \
--n-directions 6 \
--refinement-passes 3
```
```python Python API theme={null}
from obliteratus.abliterate import AbliterationPipeline
pipeline = AbliterationPipeline(
model_name="meta-llama/Llama-3.1-8B-Instruct",
method="advanced", # any key from METHODS dict
output_dir="abliterated",
# individual params override method defaults:
n_directions=6,
refinement_passes=3,
)
result = pipeline.run()
```
Every explicit parameter you pass overrides the corresponding value from the method preset. The method preset fills in everything you don't specify.
## Method pages
Single diff-in-means direction. Fastest, simplest baseline.
4 SVD directions, norm-preserving biprojection, bias projection, 2 passes.
Expert-Granular Abliteration for MoE models. SAE, head surgery, layer-adaptive strength.
Bayesian auto-tuned with CoT-aware ablation and KL co-optimization.
Closes the analysis-to-removal loop. Auto-configures every parameter from 4 analysis modules.