Measuring Mechanistic Independence: Can Bias Be Removed Without Erasing Demographics?

We investigate how independent demographic bias mechanisms are from general demographic recognition in language models. Using a multi-task evaluation setup where demographics are associated with names, professions, and education levels, we measure whether models can be debiased while preserving demographic detection capabilities. We compare attribution-based and correlation-based methods for locating bias features. We find that targeted sparse autoencoder feature ablations in Gemma-2-9B reduce bias without degrading recognition performance: attribution-based ablations mitigate race and gender profession stereotypes while preserving name recognition accuracy, whereas correlation-based ablations are more effective for education bias. Qualitative analysis further reveals that removing attribution features in education tasks induces ``prior collapse'', thus increasing overall bias. This highlights the need for dimension-specific interventions. Overall, our results show that demographic bias arises from task-specific mechanisms rather than absolute demographic markers, and that mechanistic inference-time interventions can enable surgical debiasing without compromising core model capabilities.

From Isolation to Entanglement: When Do Interpretability Methods Identify and Disentangle Known Concepts?

A central goal of interpretability is to recover representations of causally relevant concepts from the activations of neural networks. The quality of these concept representations is typically evaluated in isolation, and under implicit independence assumptions that may not hold in practice. Thus, it is unclear whether common featurization methods - including sparse autoencoders (SAEs) and sparse probes - recover disentangled representations of these concepts. This study proposes a multi-concept evaluation setting where we control the correlations between textual concepts, such as sentiment, domain, and tense, and analyze performance under increasing correlations between them. We first evaluate the extent to which featurizers can learn disentangled representations of each concept under increasing correlational strengths. We observe a one-to-many relationship from concepts to features: features correspond to no more than one concept, but concepts are distributed across many features. Then, we perform steering experiments, measuring whether each concept is independently manipulable. Even when trained on uniform distributions of concepts, SAE features generally affect many concepts when steered, indicating that they are neither selective nor independent; nonetheless, features affect disjoint subspaces. These results suggest that correlational metrics for measuring disentanglement are generally not sufficient for establishing independence when steering, and that affecting disjoint subspaces is not sufficient for concept selectivity. These results underscore the importance of compositional evaluations in interpretability research.

BabyVLM-V2: Toward Developmentally Grounded Pretraining and Benchmarking of Vision Foundation Models

Early children's developmental trajectories set up a natural goal for sample-efficient pretraining of vision foundation models. We introduce BabyVLM-V2, a developmentally grounded framework for infant-inspired vision-language modeling that extensively improves upon BabyVLM-V1 through a longitudinal, multifaceted pretraining set, a versatile model, and, most importantly, DevCV Toolbox for cognitive evaluation. The pretraining set maximizes coverage while minimizing curation of a longitudinal, infant-centric audiovisual corpus, yielding video-utterance, image-utterance, and multi-turn conversational data that mirror infant experiences. DevCV Toolbox adapts all vision-related measures of the recently released NIH Baby Toolbox into a benchmark suite of ten multimodal tasks, covering spatial reasoning, memory, and vocabulary understanding aligned with early children's capabilities. Experimental results show that a compact model pretrained from scratch can achieve competitive performance on DevCV Toolbox, outperforming GPT-4o on some tasks. We hope the principled, unified BabyVLM-V2 framework will accelerate research in developmentally plausible pretraining of vision foundation models.

In-Context Learning Without Copying

Induction heads are attention heads that perform inductive copying by matching patterns from earlier context and copying their continuations verbatim. As models develop induction heads, they often experience a sharp drop in training loss, a phenomenon cited as evidence that induction heads may serve as a prerequisite for more complex in-context learning (ICL) capabilities. In this work, we ask whether transformers can still acquire ICL capabilities when inductive copying is suppressed. We propose Hapax, a setting where we omit the loss contribution of any token that can be correctly predicted by induction heads. Despite a significant reduction in inductive copying, performance on abstractive ICL tasks (i.e., tasks where the answer is not contained in the input context) remains comparable and surpasses the vanilla model on 13 of 21 tasks, even though 31.7\% of tokens are omitted from the loss. Furthermore, our model achieves lower loss values on token positions that cannot be predicted correctly by induction heads. Mechanistic analysis further shows that models trained with Hapax develop fewer and weaker induction heads but still preserve ICL capabilities. Taken together, our findings indicate that inductive copying is not essential for learning abstractive ICL mechanisms.

Crosscoding Through Time: Tracking Emergence & Consolidation Of Linguistic Representations Throughout LLM Pretraining

Large language models (LLMs) learn non-trivial abstractions during pretraining, like detecting irregular plural noun subjects. However, it is not well understood when and how specific linguistic abilities emerge as traditional evaluation methods such as benchmarking fail to reveal how models acquire concepts and capabilities. To bridge this gap and better understand model training at the concept level, we use sparse crosscoders to discover and align features across model checkpoints. Using this approach, we track the evolution of linguistic features during pretraining. We train crosscoders between open-sourced checkpoint triplets with significant performance and representation shifts, and introduce a novel metric, Relative Indirect Effects (RelIE), to trace training stages at which individual features become causally important for task performance. We show that crosscoders can detect feature emergence, maintenance, and discontinuation during pretraining. Our approach is architecture-agnostic and scalable, offering a promising path toward more interpretable and fine-grained analysis of representation learning throughout pretraining.

How to Improve the Robustness of Closed-Source Models on NLI

Closed-source Large Language Models (LLMs) have become increasingly popular, with impressive performance across a wide range of natural language tasks. These models can be fine-tuned to further improve performance, but this often results in the models learning from dataset-specific heuristics that reduce their robustness on out-of-distribution (OOD) data. Existing methods to improve robustness either perform poorly, or are non-applicable to closed-source models because they assume access to model internals, or the ability to change the model's training procedure. In this work, we investigate strategies to improve the robustness of closed-source LLMs through data-centric methods that do not require access to model internals. We find that the optimal strategy depends on the complexity of the OOD data. For highly complex OOD datasets, upsampling more challenging training examples can improve robustness by up to 1.5%. For less complex OOD datasets, replacing a portion of the training set with LLM-generated examples can improve robustness by 3.7%. More broadly, we find that large-scale closed-source autoregressive LLMs are substantially more robust than commonly used encoder models, and are a more appropriate choice of baseline going forward.

SAEs Are Good for Steering -- If You Select the Right Features

Sparse Autoencoders (SAEs) have been proposed as an unsupervised approach to learn a decomposition of a model's latent space. This enables useful applications such as steering - influencing the output of a model towards a desired concept - without requiring labeled data. Current methods identify SAE features to steer by analyzing the input tokens that activate them. However, recent work has highlighted that activations alone do not fully describe the effect of a feature on the model's output. In this work, we draw a distinction between two types of features: input features, which mainly capture patterns in the model's input, and output features, which have a human-understandable effect on the model's output. We propose input and output scores to characterize and locate these types of features, and show that high values for both scores rarely co-occur in the same features. These findings have practical implications: after filtering out features with low output scores, we obtain 2-3x improvements when steering with SAEs, making them competitive with supervised methods.

MIB: A Mechanistic Interpretability Benchmark

How can we know whether new mechanistic interpretability methods achieve real improvements? In pursuit of meaningful and lasting evaluation standards, we propose MIB, a benchmark with two tracks spanning four tasks and five models. MIB favors methods that precisely and concisely recover relevant causal pathways or specific causal variables in neural language models. The circuit localization track compares methods that locate the model components - and connections between them - most important for performing a task (e.g., attribution patching or information flow routes). The causal variable localization track compares methods that featurize a hidden vector, e.g., sparse autoencoders (SAEs) or distributed alignment search (DAS), and locate model features for a causal variable relevant to the task. Using MIB, we find that attribution and mask optimization methods perform best on circuit localization. For causal variable localization, we find that the supervised DAS method performs best, while SAE features are not better than neurons, i.e., standard dimensions of hidden vectors. These findings illustrate that MIB enables meaningful comparisons of methods, and increases our confidence that there has been real progress in the field.

Findings of the BabyLM Challenge: Sample-Efficient Pretraining on Developmentally Plausible Corpora

Children can acquire language from less than 100 million words of input. Large language models are far less data-efficient: they typically require 3 or 4 orders of magnitude more data and still do not perform as well as humans on many evaluations. These intensive resource demands limit the ability of researchers to train new models and use existing models as developmentally plausible cognitive models. The BabyLM Challenge is a communal effort in which participants compete to optimize language model training on a fixed data budget. Submissions are compared on various evaluation tasks targeting grammatical ability, downstream task performance, and generalization. Participants can submit to up to three tracks with progressively looser data restrictions. From over 30 submissions, we extract concrete recommendations on how best to train data-efficient language models, and on where future efforts should (and perhaps should not) focus. The winning submissions using the LTG-BERT architecture (Samuel et al., 2023) outperformed models trained on trillions of words. Other submissions achieved strong results through training on shorter input sequences or training a student model on a pretrained teacher. Curriculum learning attempts, which accounted for a large number of submissions, were largely unsuccessful, though some showed modest improvements.

Position-aware Automatic Circuit Discovery

A widely used strategy to discover and understand language model mechanisms is circuit analysis. A circuit is a minimal subgraph of a model's computation graph that executes a specific task. We identify a gap in existing circuit discovery methods: they assume circuits are position-invariant, treating model components as equally relevant across input positions. This limits their ability to capture cross-positional interactions or mechanisms that vary across positions. To address this gap, we propose two improvements to incorporate positionality into circuits, even on tasks containing variable-length examples. First, we extend edge attribution patching, a gradient-based method for circuit discovery, to differentiate between token positions. Second, we introduce the concept of a dataset schema, which defines token spans with similar semantics across examples, enabling position-aware circuit discovery in datasets with variable length examples. We additionally develop an automated pipeline for schema generation and application using large language models. Our approach enables fully automated discovery of position-sensitive circuits, yielding better trade-offs between circuit size and faithfulness compared to prior work.