Language models (LMs) are susceptible to bias, sycophancy, backdoors, and other tendencies that lead to unfaithful responses to the input context. Interpreting internal states of LMs could help monitor and correct unfaithful behavior. We hypothesize that LMs faithfully represent their input contexts in a latent world model, and we seek to extract these latent world states as logical propositions. For example, given the input context ``Greg is a nurse. Laura is a physicist.'', we aim to decode the propositions WorksAs(Greg, nurse) and WorksAs(Laura, physicist) from the model's internal activations. To do so we introduce _propositional probes_, which compositionally extract lexical concepts from token activations and bind them into propositions. Key to this is identifying a _binding subspace_ in which bound tokens have high similarity (Greg $\leftrightarrow$ nurse) but unbound ones do not (Greg $\not\leftrightarrow$ physicist). Despite only being trained on linguistically simple English templates, we find that propositional probes generalize to inputs written as short stories and translated to Spanish. Moreover, in three settings where LMs respond unfaithfully to the input context---prompt injections, backdoor attacks, and gender bias--- the decoded propositions remain faithful. This suggests that LMs often encode a faithful world model but decode it unfaithfully, which motivates the search for better interpretability tools for monitoring LMs.

Pretrained language models (LMs) can generalize to implications of facts that they are finetuned on. For example, if finetuned on "John Doe lives in Tokyo," LMs correctly answer "What language do the people in John Doe’s city speak?” with "Japanese”. However, little is known about the mechanisms that enable this generalization or how they are learned during pretraining. We introduce extractive structures as a framework for describing how components in LMs (e.g., MLPs or attention heads) coordinate to enable this generalization. The structures consist of informative components that store training facts as weight changes, and upstream and downstream extractive components that query and process the stored information to produce the correct implication. We hypothesize that extractive structures are learned during pretraining when encountering implications of previously known facts. This yields two predictions: a data ordering effect where extractive structures can be learned only if facts precede their implications, and a weight grafting effect where extractive structures can be grafted to predict counterfactual implications. We empirically show these effects in the OLMo-7b, Llama 3-8b, Gemma 2-9b, and Qwen 2-7b models. Of independent interest, our results also indicate that fact learning can occur at both early and late layers, which lead to different forms of generalization.

Recent methods in language model interpretability employ techniques such as sparse autoencoders to decompose residual stream contributions into linear, semantically-meaningful features. Our work demonstrates that an underlying assumption of these methods—that residual stream contributions build additively upon each other—is insufficient to fully explain model behavior. Specifically, we identify the Transformer Layer Correction Mechanism (TLCM), wherein adjacent transformer layers systematically counteract each other's contributions to the residual stream. TLCM appears in 5 out of 7 major open-source model families and activates across nearly all tokens in diverse texts. To understand TLCM, we show that it emerges during pretraining, operates most strongly on punctuation and numbers, and adaptively calibrates its correction strength based on the preceding layer's output. We further show that TLCM actively corrects a small subspace and promotes other subspaces, different from standard model behavior. We advance the ``propose-and-reject'' hypothesis: layers may propose multiple candidate features, while subsequent layers selectively filter out inappropriate ones. Finally, we discuss how our findings help explain three persistent challenges in feature-based interpretability: why extracted features descriptions often suffer from low specificity; why feature-based interventions for model steering fail at low magnitude; why recent work finds cross-layer transcoders outperform SAEs.

Language models (LMs) can recall facts mentioned in context, as shown by their performance on reading comprehension tasks. When the context describes facts about more than one entity, the LM has to correctly bind attributes to their corresponding entity. We show, via causal experiments, that LMs' internal activations represent binding information by exhibiting appropriate binding ID vectors at the entity and attribute positions. We further show that binding ID vectors form a subspace and often transfer across tasks. Our results demonstrate that LMs learn interpretable strategies for representing symbolic knowledge in context, and that studying context activations is a fruitful direction for understanding LM cognition.

Effective planning in the real world requires not only world knowledge, but the ability to leverage that knowledge to build the right representation of the task at hand. Decades of hierarchical planning techniques have used domain-specific temporal action abstractions to support efficient and accurate planning, almost always relying on human priors and domain knowledge to decompose hard tasks into smaller subproblems appropriate for a goal or set of goals. This paper describes Ada (Action Domain Acquisition), a framework for automatically constructing task-specific planning representations using task-general background knowledge from language models (LMs). Starting with a general-purpose hierarchical planner and a low-level goal-conditioned policy, Ada interactively learns a library of planner-compatible high-level action abstractions and low-level controllers adapted to a particular domain of planning tasks. On two language-guided interactive planning benchmarks (Mini Minecraft and ALFRED Household Tasks), Ada strongly outperforms other approaches that use LMs for sequential decision-making, offering more accurate plans and better generalization to complex tasks.

We present an improved method for symbolic regression that seeks to fit data to formulas that are Pareto-optimal, in the sense of having the best accuracy for a given complexity. It improves on the previous state-of-the-art by typically being orders of magnitude more robust toward noise and bad data, and also by discovering many formulas that stumped previous methods. We develop a method for discovering generalized symmetries (arbitrary modularity in the computational graph of a formula) from gradient properties of a neural network fit. We use normalizing flows to generalize our symbolic regression method to probability distributions from which we only have samples, and employ statistical hypothesis testing to accelerate robust brute-force search.