[2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviors
Abstract:When model developers or users fine-tune an LLM, this can induce behaviors that are unexpected, deliberately harmful, or hard to detect. It would be far easier to audit LLMs if they could simply describe their behaviors in natural language. Here, we study a scalable approach to rapidly identify learned behaviors of many LLMs derived from a shared base LLM. Given a model $M$, our method works by finetuning models $M_i$ from $M$ with implanted behaviors $b_i$; the $(M_i, b_i)$ pairs serve as labeled training data. We then train an introspection adapter (IA): a single LoRA adapter jointly trained across the finetunes $M_i$ to cause them to verbalize their implanted behaviors. We find that this IA induces self-description of learned behaviors even in finetunes of $M$ that were trained in very different ways from the $M_i$. For example, IAs generalize to AuditBench, achieving state-of-the-art at identifying explicitly hidden concerning behaviors. IAs can also be used to detect encrypted finetuning API attacks. They scale favorably with model size and training data diversity. Overall, our results suggest that IAs are a scalable, effective, and practically useful approach to auditing fine-tuned LLMs.
View PDF HTML (experimental) Abstract:When model developers or users fine-tune an LLM, this can induce behaviors that are unexpected, deliberately harmful, or hard to detect. It would be far easier to audit LLMs if they could simply describe their behaviors in natural language. Here, we study a scalable approach to rapidly identify learned behaviors of many LLMs derived from a shared base LLM. Given a model $M$, our method works by finetuning models $M_i$ from $M$ with implanted behaviors $b_i$; the $(M_i, b_i)$ pairs serve as labeled training data. We then train an introspection adapter…
saved by
related reading
- [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
- Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
- Student Projects - CS 2881R AI Safetyboazbk.github.io
- Modifying LLM Beliefs with Synthetic Document Finetuningalignment.anthropic.com
- [2607.14111] Introspection Fine-Tuning (IFT): Training Small LLMs to Introspectarxiv.org
- Prefill awareness: can LLMs tell when “their” message history has been tampered with? — LessWronglesswrong.com
- Emergent introspective awareness in large language models \ Anthropicanthropic.com
- [2501.11120] Tell me about yourself: LLMs are aware of their learned behaviorsarxiv.org
- [2401.05566] Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingarxiv.org
- Discovering Language Model Behaviors with Model-Written Evaluations — LessWronglesswrong.com
- You Can Catch Sleeper Agents by Teaching Another Model to Imitate Them — LessWronglesswrong.com
- Mechanistically Eliciting Latent Behaviors in Language Models — AI Alignment Forumalignmentforum.org