# Same Outcome, Different Readout: What Does a Steerable Valence Direction in LLMs Represent?

arXiv: https://arxiv.org/abs/2609.22850 · HTML: https://mlsanity.com/paper/2609.22850/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 30 references that are in the map.
- First posted on arXiv 19 Sep 2026.

## Builds on (most cited first, 10 of 10)

- [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)
- [Designing and Interpreting Probes with Control Tasks](https://mlsanity.com/paper/1909.03368/) (2019)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [Probing the Probing Paradigm: Does Probing Accuracy Entail Task Relevance?](https://mlsanity.com/paper/2005.00719/) (2020)
- [Is This the Subspace You Are Looking for? An Interpretability Illusion for Subspace Activation Patching](https://mlsanity.com/paper/2311.17030/) (2023)
- [Towards Evaluating AI Systems for Moral Status Using Self-Reports](https://mlsanity.com/paper/2311.08576/) (2023)
- [How's it going? Reinforcement learning in language models recruits a functional welfare axis](https://mlsanity.com/paper/2605.30232/) (2026)
- [Sparse Reward Subsystem in Large Language Models](https://mlsanity.com/paper/2602.00986/) (2026)
- [Functional Emotions or Situational Contexts? A Discriminating Test from the Mythos Preview System Card](https://mlsanity.com/paper/2604.13466/) (2026)

## Abstract (arXiv)

Claims about what an internal direction in an LLM represents need evidential constraints beyond an observer's prior beliefs about the system. Decodability and successful activation steering do not, by themselves, establish which construct the direction tracks. This gap is especially consequential for welfare-relevant interpretations, where a proposed functional state must be distinguished from correlated features of the extraction contrast. We treat the question as one of construct validity and study a good-bad outcome direction in a maze task, using controlled interventions that separate the realised outcome from the informational history through which it became known. Across multiple LLM checkpoints, directions fitted on one explicit outcome encoding transfer well to another, indicating that the readout is not tied to surface form. When the same realised outcome is reached through announced and unannounced histories, however, transfer degrades substantially: even after both histories receive the same explicit outcome, the post-event readout remains strongly conditioned on the earlier announcement. In the base model, steering along the direction changes actions, yet removing it leaves the natural cue effect almost intact. In a matched maze-RL run, the post-RL direction becomes substantially more predictive of reference-MDP remaining return and the policy becomes more dependent on it at the tested sites, while the history dependence persists. These dissociations support a functional, value-related interpretation of the direction, but not its identification with a history-invariant scalar valence state.