{"ID":5443892,"CreatedAt":"2026-07-01T02:07:11.383974684Z","UpdatedAt":"2026-07-03T17:12:03.69683831Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.32008","arxiv_id":"2606.32008","title":"Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?","abstract":"Mechanistic interpretability (MI) requires full access to model internals, yet the APIs for most widely deployed language models at best expose log-probabilities over output tokens. This creates a surrogate problem: when do measurements made on open models allow us to make claims about a closed model? We evaluate surrogate fidelity at the prediction, attribution, and representation levels. For binary classification tasks, log-odds provide an API-compatible scalar readout of the model's representation space, and leave-one-out attributions provide insight into model behavior. Across eleven models spanning four families (Llama, Qwen, GPT, and Gemini), we find that prediction fidelity substantially overstates attribution fidelity: models that agree on what the answer is often disagree on why. We document an access-validity inversion: white-box signals like attention patterns and perturbation magnitudes are highly stable across models but only weakly predictive of causal attributions, which black-box input ablations capture by design. Mechanistic insight does not automatically transfer to closed targets, and prediction-level agreement is insufficient to warrant such transfer. Code and results are available at https://github.com/facebookresearch/surrogate.","short_abstract":"Mechanistic interpretability (MI) requires full access to model internals, yet the APIs for most widely deployed language models at best expose log-probabilities over output tokens. This creates a surrogate problem: when do measurements made on open models allow us to make claims about a closed model? We evaluate surro...","url_abs":"https://arxiv.org/abs/2606.32008","url_pdf":"https://arxiv.org/pdf/2606.32008v1","authors":"[\"Philippe Chlenski\",\"Zachariah Carmichael\",\"Ayush Warikoo\",\"Chia-Tse Shao\",\"Yingxiao Ye\",\"Aobo Yang\",\"Vivek Miglani\",\"Nehal Bandi\"]","published":"2026-06-30T17:43:44Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false,"code_links":[{"ID":613823,"CreatedAt":"2026-07-01T02:07:11.383974684Z","UpdatedAt":"2026-07-01T02:07:11.383974684Z","DeletedAt":null,"paper_id":5443892,"paper_url":"https://arxiv.org/abs/2606.32008","paper_title":"Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?","repo_url":"https://github.com/facebookresearch/surrogate","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}