# Steering off Course: Reliability Challenges in Steering Language Models

arXiv: https://arxiv.org/abs/2504.04635 · HTML: https://mlsanity.com/paper/2504.04635/

## Key facts

- Cited by 34 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 69 references that are in the map.
- First posted on arXiv 6 Apr 2025.

## Builds on (most cited first, 30 of 40)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [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)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [In-context Learning and Induction Heads](https://mlsanity.com/paper/2209.11895/) (2022)
- [Towards Automated Circuit Discovery for Mechanistic Interpretability](https://mlsanity.com/paper/2304.14997/) (2023)
- [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)
- [Quantifying the Carbon Emissions of Machine Learning](https://mlsanity.com/paper/1910.09700/) (2019)
- [Causal Abstractions of Neural Networks](https://mlsanity.com/paper/2106.02997/) (2021)
- [Editing Large Language Models: Problems, Methods, and Opportunities](https://mlsanity.com/paper/2305.13172/) (2023)
- [Finding Neurons in a Haystack: Case Studies with Sparse Probing](https://mlsanity.com/paper/2305.01610/) (2023)
- [Emergent Linear Representations in World Models of Self-Supervised Sequence Models](https://mlsanity.com/paper/2309.00941/) (2023)
- [In-Context Learning Creates Task Vectors](https://mlsanity.com/paper/2310.15916/) (2023)
- [Fine-tuning Language Models for Factuality](https://mlsanity.com/paper/2311.08401/) (2023)
- [AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders](https://mlsanity.com/paper/2501.17148/) (2025)
- [Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias](https://mlsanity.com/paper/2004.12265/) (2020)
- [Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla](https://mlsanity.com/paper/2307.09458/) (2023)
- [How to Train Long-Context Language Models (Effectively)](https://mlsanity.com/paper/2410.02660/) (2024)
- [Analyzing the Generalization and Reliability of Steering Vectors](https://mlsanity.com/paper/2407.12404/) (2024)
- [Devils in Middle Layers of Large Vision-Language Models: Interpreting, Detecting and Mitigating Object Hallucinations via Attention Lens](https://mlsanity.com/paper/2411.16724/) (2024)
- [Generating Benchmarks for Factuality Evaluation of Language Models](https://mlsanity.com/paper/2307.06908/) (2023)

## Built on by (most cited first, 30 of 34)

- [Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models](https://mlsanity.com/paper/2601.14004/) (2026)
- [Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior](https://mlsanity.com/paper/2605.05115/) (2026)
- [KV Cache Steering for Controlling Frozen LLMs](https://mlsanity.com/paper/2507.08799/) (2025)
- [Analysing the Safety Pitfalls of Steering Vectors](https://mlsanity.com/paper/2603.24543/) (2026)
- [Steered LLM Activations are Non-Surjective](https://mlsanity.com/paper/2604.09839/) (2026)
- [EasyEdit2: An Easy-to-use Steering Framework for Editing Large Language Models](https://mlsanity.com/paper/2504.15133/) (2025)
- [IF-GUIDE: Influence Function-Guided Detoxification of LLMs](https://mlsanity.com/paper/2506.01790/) (2025)
- [What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal](https://mlsanity.com/paper/2604.08524/) (2026)
- [Steerable but Not Decodable: Function Vectors Operate Beyond the Logit Lens](https://mlsanity.com/paper/2604.02608/) (2026)
- [Activation-Space Personality Steering: Hybrid Layer Selection for Stable Trait Control in LLMs](https://mlsanity.com/paper/2511.03738/) (2025)
- [Compositional Steering of Large Language Models with Steering Tokens](https://mlsanity.com/paper/2601.05062/) (2026)
- [The Effectiveness of Style Vectors for Steering Large Language Models: A Human Evaluation](https://mlsanity.com/paper/2601.21505/) (2026)
- [Beyond representational alignment with brain-guided language models for robust reasoning](https://mlsanity.com/paper/2606.11893/) (2026)
- [ThoughtProbe: Classifier-Guided LLM Thought Space Exploration via Probing Representations](https://mlsanity.com/paper/2510.27355/) (2025)
- [Eliciting Chain-of-Thought in Base LLMs via Gradient-Based Representation Optimization](https://mlsanity.com/paper/2511.19131/) (2025)
- [Neuron-Level Emotion Control in Speech-Generative Large Audio-Language Models](https://mlsanity.com/paper/2603.17231/) (2026)
- [Decodable but Not Corrected by Fixed Residual-Stream Linear Steering: Evidence from Medical LLM Failure Regimes](https://mlsanity.com/paper/2605.05715/) (2026)
- [Conditional Optimal Bridge for Riemannian Activation Steering](https://mlsanity.com/paper/2607.10517/) (2026)
- [IA2: Alignment with ICL Activations Improves Supervised Fine-Tuning](https://mlsanity.com/paper/2509.22621/) (2025)
- [Riemannian-Manifold Steering: Geometry-Aware Generative Autoencoders for Label-Free Steering](https://mlsanity.com/paper/2605.24942/) (2026)
- [SteerCheck: Attribution Specificity and Alignment Leakage in Activation-Steering Audits](https://mlsanity.com/paper/2608.24335/) (2026)
- [Logit-Gap Steering: A Forward-Pass Diagnostic for Alignment Robustness](https://mlsanity.com/paper/2506.24056/) (2025)
- [Facet-Level Persona Control by Trait-Activated Routing with Contrastive SAE for Role-Playing LLMs](https://mlsanity.com/paper/2602.19157/) (2026)
- [Activation Steering for Synthetic Data Generation: The Role of Diversity in Downstream Safety Detection](https://mlsanity.com/paper/2605.28664/) (2026)
- [Sense Representations Are Inducible Interfaces](https://mlsanity.com/paper/2605.28669/) (2026)
- [Order Is Not Control: Driven-Dissipative Response Laws Across Artificial and Biological Systems](https://mlsanity.com/paper/2606.12923/) (2026)
- [Probabilistic Concept-Aware Steering for Trustworthy LLM Inference](https://mlsanity.com/paper/2607.18259/) (2026)
- [From Profiles to Steering Vectors: Global Sparse Priors and Local Semantic Calibration for Personalized Text Generation](https://mlsanity.com/paper/2607.21620/) (2026)
- [Signatures of Steerability in Activation Space of Language Models](https://mlsanity.com/paper/2609.14151/) (2026)
- ["As a Language Model...": Chat Template Switches LLM Self-Referential Voice and Activation Steering Reproduces It](https://mlsanity.com/paper/2609.25021/) (2026)

## Abstract (arXiv)

Steering methods for language models (LMs) have gained traction as lightweight alternatives to fine-tuning, enabling targeted modifications to model activations. However, prior studies primarily report results on a few models, leaving critical gaps in understanding the robustness of these methods. In this work, we systematically examine three prominent steering methods -- DoLa, function vectors, and task vectors. In contrast to the original studies, which evaluated a handful of models, we test up to 36 models belonging to 14 families with sizes ranging from 1.5B to 70B parameters. Our experiments reveal substantial variability in the effectiveness of the steering approaches, with a large number of models showing no improvement and at times degradation in steering performance. Our analysis demonstrate fundamental flaws in the assumptions underlying these methods, challenging their reliability as scalable steering solutions.