# Depth-Wise Activation Steering for Honest Language Models

arXiv: https://arxiv.org/abs/2512.07667 · HTML: https://mlsanity.com/paper/2512.07667/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Probing deception, bias and beliefs, one of 13 directions in Mechanistic interpretability.
- Builds on 14 of its 18 references that are in the map.
- First posted on arXiv 8 Dec 2025.

## Placed in

- Mechanistic interpretability / [Probing deception, bias and beliefs](https://mlsanity.com/mechinterp/probing-deception-bias-and-beliefs/)

## Builds on (most cited first, 14 of 14)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- ["Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models](https://mlsanity.com/paper/2308.03825/) (2023)
- [Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmark](https://mlsanity.com/paper/2304.03279/) (2023)
- [On Prompt-Driven Safeguarding for Large Language Models](https://mlsanity.com/paper/2401.18018/) (2024)
- [Steering Language Model Refusal with Sparse Autoencoders](https://mlsanity.com/paper/2411.11296/) (2024)
- [Improving Steering Vectors by Targeting Sparse Autoencoder Features](https://mlsanity.com/paper/2411.02193/) (2024)
- [The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems](https://mlsanity.com/paper/2503.03750/) (2025)
- [Semantics-Adaptive Activation Intervention for LLMs via Dynamic Steering Vectors](https://mlsanity.com/paper/2410.12299/) (2024)
- [Activation Scaling for Steering and Interpreting Language Models](https://mlsanity.com/paper/2410.04962/) (2024)
- [Beyond Single Concept Vector: Modeling Concept Subspace in LLMs with Gaussian Distribution](https://mlsanity.com/paper/2410.00153/) (2024)

## Built on by (most cited first, 4 of 4)

- [Efficient and accurate steering of Large Language Models through attention-guided feature learning](https://mlsanity.com/paper/2602.00333/) (2026)
- [Preconditioned Test-Time Adaptation for Out-of-Distribution Debiasing in Narrative Generation](https://mlsanity.com/paper/2603.13683/) (2026)
- [Playing Devil's Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy](https://mlsanity.com/paper/2605.21006/) (2026)
- [Auditing Asset-Specific Preferences in Financial Large Language Models: Evidence from Bitcoin Representations and Portfolio Allocation](https://mlsanity.com/paper/2606.02528/) (2026)

## Abstract (arXiv)

Large language models sometimes assert falsehoods despite internally representing the correct answer, failures of honesty rather than accuracy, which undermines auditability and safety. Existing approaches largely optimize factual correctness or depend on retraining and brittle single-layer edits, offering limited leverage over truthful reporting. We present a training-free activation steering method that weights steering strength across network depth using a Gaussian schedule. On the MASK benchmark, which separates honesty from knowledge, we evaluate seven models spanning the LLaMA, Qwen, and Mistral families and find that Gaussian scheduling improves honesty over no-steering and single-layer baselines in six of seven models. Equal-budget ablations on LLaMA-3.1-8B-Instruct and Qwen-2.5-7B-Instruct show the Gaussian schedule outperforms random, uniform, and box-filter depth allocations, indicating that how intervention is distributed across depth materially affects outcomes beyond total strength. The method is simple, model-agnostic, requires no finetuning, and provides a low-cost control knob for eliciting truthful reporting from models' existing capabilities.