# Would this change your answer? Evaluating Explanations of LLM Behavior In The Wild with Counterfactual Experiments

arXiv: https://arxiv.org/abs/2608.16747 · HTML: https://mlsanity.com/paper/2608.16747/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Causal reasoning in LLMs, one of 17 directions in LLM reasoning.
- Builds on 20 of its 31 references that are in the map.
- First posted on arXiv 17 Aug 2026.

## Placed in

- LLM reasoning / [Causal reasoning in LLMs](https://mlsanity.com/lr/causal-reasoning-in-llms/)

## Builds on (most cited first, 20 of 20)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Chain-of-Thought Reasoning In The Wild Is Not Always Faithful](https://mlsanity.com/paper/2503.08679/) (2025)
- [Looking Inward: Language Models Can Learn About Themselves by Introspection](https://mlsanity.com/paper/2410.13787/) (2024)
- [Scruples: A Corpus of Community Ethical Judgments on 32,000 Real-Life Anecdotes](https://mlsanity.com/paper/2008.09094/) (2020)
- [Tell me about yourself: LLMs are aware of their learned behaviors](https://mlsanity.com/paper/2501.11120/) (2025)
- [Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanations](https://mlsanity.com/paper/2307.08678/) (2023)
- [Bias-Augmented Consistency Training Reduces Biased Reasoning in Chain-of-Thought](https://mlsanity.com/paper/2403.05518/) (2024)
- [AuditBench: Evaluating Alignment Auditing Techniques on Models with Hidden Behaviors](https://mlsanity.com/paper/2602.22755/) (2026)
- [Training Language Models to Explain Their Own Computations](https://mlsanity.com/paper/2511.08579/) (2025)
- [Narrow Finetuning Leaves Clearly Readable Traces in Activation Differences](https://mlsanity.com/paper/2510.13900/) (2025)
- [Eliciting Secret Knowledge from Language Models](https://mlsanity.com/paper/2510.01070/) (2025)
- [Teaching Models to Verbalize Reward Hacking in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2506.22777/) (2025)
- [Counterfactual Simulation Training for Chain-of-Thought Faithfulness](https://mlsanity.com/paper/2602.20710/) (2026)
- [LLMs Don't Know Their Own Decision Boundaries: The Unreliability of Self-Generated Counterfactual Explanations](https://mlsanity.com/paper/2509.09396/) (2025)
- [A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior](https://mlsanity.com/paper/2602.02639/) (2026)
- [Can LLMs Explain Themselves Counterfactually?](https://mlsanity.com/paper/2502.18156/) (2025)
- [The Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training Methodology](https://mlsanity.com/paper/2607.01033/) (2026)
- [Introspective Coupling: Self-Explanation Training Tracks Behavioral Change Despite Fixed Supervision](https://mlsanity.com/paper/2606.32038/) (2026)

## Built on by (most cited first, 2 of 2)

- [Understanding Confabulation and Rethinking Reconstruction in Activation Explanations](https://mlsanity.com/paper/2609.33702/) (2026)
- [Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts](https://mlsanity.com/paper/2610.00314/) (2026)

## Abstract (arXiv)

Many areas of AI research, such as language model interpretability and chain of thought faithfulness, seek to explain model behaviors. But what constitutes a "good" explanation? In this work, we evaluate explanations through the lens of counterfactual simulatability-whether the explanation is useful for predicting model behaviors on related counterfactual inputs. To this end, we introduce CHIVE (Counterfactual Hypothesis Investigation Via Edits), a novel agentic pipeline that identifies unexpected model behaviors in the wild and investigates them with counterfactual prompt edits. This yields thousands of high-quality explanations for naturally-occurring model behaviors along with supporting counterfactual evidence. We apply CHIVE in two ways. First, we evaluate whether common LLM interpretability techniques improve an agent's ability to predict counterfactual model behaviors. Surprisingly, we find no uplift from any of the interpretability techniques studied. Second, we use CHIVE to generate training data. We find that training models to predict outcomes of CHIVE-generated counterfactual experiments generalizes to various out-of-distribution settings. Overall, CHIVE automatically discovers explanations of naturally-occurring LLM behaviors, enabling us to evaluate and improve methods for explaining LLM behaviors.