Small Models Can Introspect, Too

Doing introspection experiments with an OSS model.

Small Models Can Introspect, Too Posted December 12, 2025 Recent work by Anthropic showed that Claude models, primarily Opus 4 and Opus 4.1, are able to introspect--detecting when external concepts have been injected into their activations. But not all of us have Opus at home! By looking at the logits, we show that a 32B open-source model that at first appears unable to introspect actually is subtly introspecting. We then show that better prompting can significantly improve introspection performance, and throw the logit lens and emergent misalignment into the mix, showing that the model can in

related reading

• Emergent introspective awareness in large language models \ Anthropicanthropic.com
• Emergent Introspective Awareness in Large Language Modelstransformer-circuits.pub
• Emergent Introspective Awareness in Large Language Modelstransformer-circuits.pub
• Tracing the thoughts of a large language model \ Anthropicanthropic.com
• Introspection or entropy? Re-examining concept-injection “introspection” in open models — LessWronglesswrong.com
• [2607.14111] Introspection Fine-Tuning (IFT): Training Small LLMs to Introspectarxiv.org
• On the Biology of a Large Language Modeltransformer-circuits.pub
• [2602.20031] Latent Introspection: Models Can Detect Prior Concept Injectionsarxiv.org
• A global workspace in language models \ Anthropicanthropic.com
• [2601.01828] Emergent Introspective Awareness in Large Language Modelsarxiv.org
• [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• Emergent Introspection in AI is Content-Agnosticarxiv.org