# A Pre-Specified Construction-Confirmation Test of Operation-Level Causal Transfer Across Finite Isomorphic Symbolic Domains

arXiv: https://arxiv.org/abs/2608.15809 · HTML: https://mlsanity.com/paper/2608.15809/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 20 references that are in the map.
- First posted on arXiv 16 Aug 2026.

## Builds on (most cited first, 19 of 19)

- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task](https://mlsanity.com/paper/2210.13382/) (2022)
- [Causal Abstractions of Neural Networks](https://mlsanity.com/paper/2106.02997/) (2021)
- [In-Context Learning Creates Task Vectors](https://mlsanity.com/paper/2310.15916/) (2023)
- [Function Vectors in Large Language Models](https://mlsanity.com/paper/2310.15213/) (2023)
- [In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering](https://mlsanity.com/paper/2311.06668/) (2023)
- [Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations](https://mlsanity.com/paper/2303.02536/) (2023)
- [Successor Heads: Recurring, Interpretable Attention Heads In The Wild](https://mlsanity.com/paper/2312.09230/) (2023)
- [Emergent Symbolic Mechanisms Support Abstract Reasoning in Large Language Models](https://mlsanity.com/paper/2502.20332/) (2025)
- [mOthello: When Do Cross-Lingual Representation Alignment and Cross-Lingual Transfer Emerge in Multilingual Models?](https://mlsanity.com/paper/2404.12444/) (2024)
- [Causality neq Invariance: Function and Concept Vectors in LLMs](https://mlsanity.com/paper/2602.22424/) (2026)
- [AlgoTrace: Algorithmic Primitives and Compositional Geometry of Reasoning in Language Models](https://mlsanity.com/paper/2510.15987/) (2025)
- [Emergent Analogical Reasoning in Transformers](https://mlsanity.com/paper/2602.01992/) (2026)
- [MetaOthello: A Controlled Study of Multiple World Models in Transformers](https://mlsanity.com/paper/2602.23164/) (2026)
- [Transport of Algebraic Structure to Latent Embeddings](https://mlsanity.com/paper/2405.16763/) (2024)
- [Categorical Perception in Large Language Model Hidden States: Structural Warping at Digit-Count Boundaries](https://mlsanity.com/paper/2603.28258/) (2026)
- [Can Large Language Models Generalize Procedures Across Representations?](https://mlsanity.com/paper/2602.03542/) (2026)
- [Single-Position Intervention Fails: Distributed Output Templates Drive In-Context Learning](https://mlsanity.com/paper/2605.04061/) (2026)
- [Are Arithmetic Heuristic Neurons Form-Invariant? A Mechanistic Analysis of Symbols, Text, and Code in LLMs](https://mlsanity.com/paper/2607.16693/) (2026)

## Abstract (arXiv)

Behavioral accuracy, linear decodability, and successful activation interventions do not by themselves show that a model carries an operation-level structure from one symbolic domain to another. We ask a narrower question in finite isomorphic state spaces: if the hidden-state difference between two operations is estimated separately for each source input, does adding that difference to a mapped recipient input move the model toward the corresponding recipient answer? The design compares this input-specific intervention with wrong-operation, norm-matched random, and no-op controls, and separates candidate construction from an independently isolated confirmation split. On a frozen Qwen2.5-7B-Instruct model at layers 20--21, one route--domain--operation candidate from a family pre-specified and frozen before confirmation access, transparent | integer_mod16--letters16 | successor->predecessor, passed both PyVene splits; its confirmation intersection--union p-value was 0.000198 and its 36-family Holm-adjusted p-value was 0.006943. A subsequent NNsight 0.7.0 experiment, pre-specified and frozen before its confirmation access, tested only this selected prompt route, without candidate or layer reselection. It reproduced all 12 confirmation effect estimates, confidence intervals, and exact sign-flip p-values numerically; its 36-family Holm-adjusted p-value was 0.007141. The result is therefore limited to one prompt route and one candidate, replicated across two intervention implementations on one model revision and one layer interval. It does not establish cross-model generalization, full-family backend independence, domain-general transfer, or algebraic invariance.