Leverage Is Not Reach: A Control-Window Law for Single-Neuron Steering in Language Models

• URL: http://arxiv.org/abs/2606.19831v1
• Date: Thu, 18 Jun 2026 06:25:18 GMT
• Title: Leverage Is Not Reach: A Control-Window Law for Single-Neuron Steering in Language Models
• Abstract summary: We develop a budget normalized control window framework for single neuron steering.<n>A dose along one write direction reduces to one control coordinate.<n>Coherent control exists when a behavior trigger lies below the collapse ceiling.
• Score: 1.7259824817932294
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Aligned language models gate behaviors such as refusal and language routing through sparse feed forward neurons, yet no theory predicts when a single neuron intervention controls a behavior coherently rather than collapsing the output. We develop a budget normalized control window framework for single neuron steering. A dose along one write direction reduces to one control coordinate: the alignment between the residual stream and the write, driven along a universal saturation curve in units of a coherence budget set by the residual norm divided by the write norm. Coherent control exists when a behavior trigger lies below the collapse ceiling. The same coordinate governs benign mode switches and refusal; the ceiling follows from weights and one generic forward pass, while triggers are measured at rollout. On fifteen held out neurons, the predicted ceiling has mean absolute error 0.14, about 0.07 in bulk layers, and the committed open or closed verdict holds on eleven against a ten of fifteen majority baseline. Closed cases expose three failure modes rather than violations: collapse before trigger, too little depth to propagate, or a normalization that caps how far one neuron can push. The law explains why local gradient attribution anti predicts control: true controllers write off the readout axis and carry a near zero first order gradient. A forward only contrastive screen made precise by the window recovers controllers that attribution misses. On refusal, the hardest case, intervention success is typed, not scalar: coherent bypass and strict actionable reach separate, so a neuron can flip refusal in fluent, on task text with no actionable content, and genuine actionable reach appears only for three of six audited Llama pivots and only at later rollout horizons. Single neuron steering is therefore a budgeted, typed audit of controllability rather than a fixed dose anecdote.

Related papers

• GAPS: Dimension-Level Gates for Conditional Activation Steering [5.5832442227189425]
  Activation steering suppresses undesired behaviors in language models by adding a steering vector to the hidden state during generation.<n>Recent conditional methods such as CAST and DSAS improve the behavior-capability trade-off by deciding when to intervene, but once active, they apply the full dense vector to all hidden dimensions.<n>We introduce dimension-level conditioning as a complementary axis of selectivity that also decides which neurons to intervene on.
  arXiv  Detail & Related papers  (2026-09-01T21:21:52Z)
• Feature Priming in Online Linear Regression: Sparse-Regret Lower Bounds and a Tight Univariate Rate [47.36630149538994]
  In high-dimensional online prediction, the best predictor may depend on only a few features, so regret should scale with sparsity rather than the ambient dimension.<n>Warmuth and Amid asked at COLT 2023 whether any of three such rules admits a competitive online regret guarantee.<n>Our analysis identifies a common obstruction: cheap nuisance causes the refit to underweight the truly predictive coordinate.
  arXiv  Detail & Related papers  (2026-08-18T09:32:03Z)
• When Is a Steerable Concept Representation Real? Measurement Confounds in a Cross-Family Audit of Neuroscience Parallels in LLMs [8.657685524975246]
  Large language models (LLMs) are increasingly reported to exhibit human-like neural and cognitive signatures.<n>Our main experiment examines the causal steerability of concept directions.<n>With raw activation units and a fixed layer and coefficient, steerability appears to increase with model scale.<n>With residual-norm-comparable interventions and held-out operating-point selection, concept steering remains significant at every scale.
  arXiv  Detail & Related papers  (2026-08-08T14:37:53Z)
• No Single Neuron of Failure: Distributed Safety Alignment Against White-Box Attacks [61.07406641465909]
  We propose distributed safety alignment, which encodes safety capabilities across multiple computational neurons.<n>We show that DSA substantially improves against white-box neuron-level safety attacks while preserving the model's general language and multimodal utility.
  arXiv  Detail & Related papers  (2026-08-02T17:49:07Z)
• Minimizing Targeted Activations: Input-Only Suppression of Evaluation-Awareness Latents in Large Language Models [0.4783156204372467]
  Activation steering controls model behavior by editing internal activations at inference time.<n>We study its input-side dual: optimizing a fluent prompt so that a chosen internal latent is driven toward zero.
  arXiv  Detail & Related papers  (2026-07-28T16:01:48Z)
• Faithfulness to Refusal: A Causal Audit of Neuron Selectors [3.3598755777055374]
  Attribution scores increasingly identify which neuron rows of a language model matter for applications such as pruning, interpretability, and editing for safety.<n>We address this with two paired audits built on one-shot neuron-row zeroing.
  arXiv  Detail & Related papers  (2026-07-06T17:33:36Z)
• Output Vector Editing for Memorization Mitigation in Large Language Models [68.30351930772788]
  Large language models memorize and reproduce sequences from their training data, creating privacy, copyright, and security risks.<n>Existing neuron-level mitigation methods equate editing with zeroing out neuron activations, but the activation only controls whether a neuron engages.<n>We propose output vector editing, a constrained-optimization edit that locates a small set of neurons responsible for a memorized continuation.
  arXiv  Detail & Related papers  (2026-06-17T07:29:18Z)
• Targeted Neuron Modulation via Contrastive Pair Search [0.0]
  We show that neuron-level intervention enables reliable behavioral steering without the quality tradeoffs of residual-stream methods.<n>We find that base models contain similar late-layer discrimination structures but steering these neurons produces only content shifts.<n>More broadly, our findings suggest that alignment fine-tuning transforms pre-existing discrimination structure into a sparse, targetable refusal gate.
  arXiv  Detail & Related papers  (2026-05-12T15:47:38Z)
• Neuron-Anchored Rule Extraction for Large Language Models via Contrastive Hierarchical Ablation [5.880505093493663]
  A key goal of explainable AI (XAI) is to express the decision logic of large language models (LLMs) in symbolic form.<n>We introduce MechaRule, a pipeline that grounds rule extraction in LLM circuits by efficiently localizing sparse neurons called agonists.
  arXiv  Detail & Related papers  (2026-05-04T18:27:37Z)
• Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors [48.881343993730844]
  Safety-aligned large language models (LLMs) are increasingly deployed in real-world pipelines.<n> adversaries can distribute backdoored checkpoints that behave normally under standard evaluation but jailbreak when a hidden trigger is present.<n>Recent post-hoc weight-editing methods offer an efficient approach to injecting such backdoors.
  arXiv  Detail & Related papers  (2026-04-14T06:48:33Z)
• How Alignment Routes: Localizing, Scaling, and Controlling Policy Circuits in Language Models [0.0]
  This paper localizes the policy routing mechanism in alignment-trained language models.<n>An intermediate-layer attention gate reads detected content and triggers deeper heads that boost the signal toward refusal.
  arXiv  Detail & Related papers  (2026-04-06T03:20:37Z)
• Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails [0.0]
  We study political censorship in Chinese-origin language models as a natural experiment.<n>We use probes, surgical ablations, and behavioral tests across nine open-weight models from five labs.
  arXiv  Detail & Related papers  (2026-03-18T20:54:34Z)
• On the Structural Non-Preservation of Epistemic Behaviour under Policy Transformation [51.56484100374058]
  We formalise such information-conditioned interaction patterns as behavioural dependency.<n>This induces a probe-relative notion of $$-behavioural equivalence and a within-policy behavioural distance.<n>Results identify structural conditions under which probe-conditioned behavioural separation is not preserved under common policy transformations.
  arXiv  Detail & Related papers  (2026-02-24T22:55:21Z)
• Verifying Closed-Loop Contractivity of Learning-Based Controllers via Partitioning [52.23804865017831]
  We address the problem of verifying closed-loop contraction in nonlinear control systems whose controller and contraction metric are both parameterized by neural networks.<n>We derive a tractable and scalable sufficient condition for closed-loop contractivity that reduces to checking that the dominant eigenvalue of a symmetric Metzler matrix is nonpositive.
  arXiv  Detail & Related papers  (2025-12-01T23:06:56Z)
• Population Dynamics Control with Partial Observations [4.753557469026313]
  We study the problem of controlling population dynamics from the perspective of online non-stochastic control.<n>Our main contribution is a new controller that achieves the optimal $tildeO(sqrtT)$ regret with respect to a natural class of mixing linear dynamic controllers.
  arXiv  Detail & Related papers  (2025-02-19T20:07:56Z)
• Neuro-Symbolic Entropy Regularization [78.16196949641079]
  In structured prediction, the goal is to jointly predict many output variables that together encode a structured object. One approach -- entropy regularization -- posits that decision boundaries should lie in low-probability regions. We propose a loss, neuro-symbolic entropy regularization, that encourages the model to confidently predict a valid object.
  arXiv  Detail & Related papers  (2022-01-25T06:23:10Z)
• Characterizing and addressing the issue of oversmoothing in neural autoregressive sequence modeling [49.06391831200667]
  We study the effect of the proposed regularization on both model distribution and decoding performance. We conclude that the high degree of oversmoothing is the main reason behind the case of overly probable short sequences in a neural autoregressive model.
  arXiv  Detail & Related papers  (2021-12-16T14:33:12Z)
• And/or trade-off in artificial neurons: impact on adversarial robustness [91.3755431537592]
  Presence of sufficient number of OR-like neurons in a network can lead to classification brittleness and increased vulnerability to adversarial attacks. We define AND-like neurons and propose measures to increase their proportion in the network. Experimental results on the MNIST dataset suggest that our approach holds promise as a direction for further exploration.
  arXiv  Detail & Related papers  (2021-02-15T08:19:05Z)
• Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-Level Backdoor Attacks [98.15243373574518]
  Pre-trained models (PTMs) have been widely used in various downstream tasks. In this work, we demonstrate the universal vulnerability of PTMs, where fine-tuned PTMs can be easily controlled by backdoor attacks.
  arXiv  Detail & Related papers  (2021-01-18T10:18:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.