✳flâneur — a map of the web's best reading

[2607.14111] Introspection Fine-Tuning (IFT): Training Small LLMs to Introspect

arxiv.org · saved by 1 readers

Abstract:Can small language models detect and report on perturbations their own internal activations? We investigate this question through the lens of activation steering: injecting concept vectors into a model's residual stream and measuring whether the model can accurately report on the perturbation. We first show that the binary detection paradigm used in prior work -- prompting the model to answer Yes'' or No'' to whether it detects an injected thought -- is confounded in small models, as steering biases the model toward affirmative responses regardless of the question content. We therefore propose two confound-free evaluation paradigms: sentence localization (identifying which of $N$ sentences was perturbed, chance $= 1/N$) and strength comparison (identifying which of two sentences received a stronger injection, chance $= 50\%$). Evaluating across six models from two families (Llama-3.2 and Gemma-4), we find that models as small as 2B parameters introspect reliably well above chance, and that introspective ability generally increases with scale. Llama-1B, however, performs at or below chance. We then introduce \emph{Introspection Fine-Tuning} (IFT): supervised fine-tuning on sentence-localization examples constructed from the model's own perturbed forward passes. IFT raises Llama-1B sentence-localization accuracy from $9.6\%$ to $60.6\%$ (a $6\times$ improvement), with gains generalizing zero-shot to the held-out strength-comparison task ($30.2\% \to 52.2\%$). IFT also improves introspection for 3B and 8B models, while inducing negligible degradation on standard capability benchmarks. Our results suggest that introspective ability is not fixed by scale alone: it can be directly trained, and doing so unlocks latent self-monitoring capacity with implications for AI transparency and alignment. Our code is \href{this https URL}{here}.

Abstract:Can small language models detect and report on perturbations their own internal activations? We investigate this question through the lens of activation steering: injecting concept vectors into a model's residual stream and measuring whether the model can accurately report on the perturbation. We first show that the binary detection paradigm used in prior work -- prompting the model to answer Yes'' or No'' to whether it detects an injected thought -- is confounded in small models, as steering biases the model toward affirmative responses regardless of the question content. We therefor

Explore this link on the map →

related reading