# Precise In-Parameter Concept Erasure in Large Language Models

arXiv: https://arxiv.org/abs/2505.22586 · HTML: https://mlsanity.com/paper/2505.22586/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 59 of its 87 references that are in the map.
- First posted on arXiv 28 May 2025.

## Builds on (most cited first, 30 of 59)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [How Much Knowledge Can You Pack Into the Parameters of a Language Model?](https://mlsanity.com/paper/2002.08910/) (2020)
- [Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space](https://mlsanity.com/paper/2203.14680/) (2022)
- [Dissecting Recall of Factual Associations in Auto-Regressive Language Models](https://mlsanity.com/paper/2304.14767/) (2023)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [Making AI Forget You: Data Deletion in Machine Learning](https://mlsanity.com/paper/1907.05012/) (2019)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models](https://mlsanity.com/paper/2403.19647/) (2024)
- [Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2](https://mlsanity.com/paper/2408.05147/) (2024)
- [Who's Harry Potter? Approximate Unlearning in LLMs](https://mlsanity.com/paper/2310.02238/) (2023)
- [Pay Attention to MLPs](https://mlsanity.com/paper/2105.08050/) (2021)
- [Finding Neurons in a Haystack: Case Studies with Sparse Probing](https://mlsanity.com/paper/2305.01610/) (2023)
- [Improving Alignment and Robustness with Circuit Breakers](https://mlsanity.com/paper/2406.04313/) (2024)
- [LEACE: Perfect linear concept erasure in closed form](https://mlsanity.com/paper/2306.03819/) (2023)
- [AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders](https://mlsanity.com/paper/2501.17148/) (2025)

## Built on by (most cited first, 11 of 11)

- [Watch the Weights: Unsupervised monitoring and control of fine-tuned LLMs](https://mlsanity.com/paper/2508.00161/) (2025)
- [Interpretability Can Be Actionable](https://mlsanity.com/paper/2605.11161/) (2026)
- [Transmuting prompts into weights](https://mlsanity.com/paper/2510.08734/) (2025)
- [Disentangling MLP Neuron Weights in Vocabulary Space](https://mlsanity.com/paper/2604.06005/) (2026)
- [From Memorization to Reasoning in the Spectrum of Loss Curvature](https://mlsanity.com/paper/2510.24256/) (2025)
- [Steerability of Instrumental-Convergence Tendencies in LLMs](https://mlsanity.com/paper/2601.01584/) (2026)
- [FIT to Forget: Robust Continual Unlearning for Large Language Models](https://mlsanity.com/paper/2601.21682/) (2026)
- [Understanding Empirical Unlearning with Combinatorial Interpretability](https://mlsanity.com/paper/2602.19215/) (2026)
- [Safety Anchor: Defending Harmful Fine-tuning via Geometric Bottlenecks](https://mlsanity.com/paper/2605.05995/) (2026)
- [Reinforcement Learning for Neural Model Editing](https://mlsanity.com/paper/2606.13461/) (2026)
- [Weights Read and Write Features: Scalable Parameter Decomposition Grounded in Activation Space](https://mlsanity.com/paper/2609.37731/) (2026)

## Abstract (arXiv)

Large language models (LLMs) often acquire knowledge during pretraining that is undesirable in downstream deployments, e.g., sensitive information or copyrighted content. Existing approaches for removing such knowledge rely on fine-tuning, training low-rank adapters or fact-level editing, but these are either too coarse, too shallow, or ineffective. In this work, we propose PISCES (Precise In-parameter Suppression for Concept EraSure), a novel framework for precisely erasing entire concepts from model parameters by directly editing directions that encode them in parameter space. PISCES uses a disentangler model to decompose MLP vectors into interpretable features, identifies those associated with a target concept using automated interpretability techniques, and removes them from model parameters. Experiments on Gemma 2 and Llama 3.1 over various concepts show that PISCES achieves modest gains in efficacy over leading erasure methods, reducing accuracy on the target concept to as low as 7.7%, while dramatically improving erasure specificity (by up to 31%) and robustness (by up to 38%). Overall, these results demonstrate that feature-based in-parameter editing enables a more precise and reliable approach for removing conceptual knowledge in language models.