SFAL: Semantic-Functional Alignment Scores for Distributional Evaluation of Auto-Interpretability in Sparse Autoencoders (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Interpreting the internal representations of large language models (LLMs) is crucial for their deployment in real-world applications, impacting areas such as AI safety, debugging, and compliance. |
| Approach: | They propose an alternative evaluation strategy that assesses the alignment between the semantic neighbourhoods of features and their functional neighbourhoods by using co-occurrence statistics. |
| Outcome: | The proposed evaluation strategy reduces reliance on scoring on large-scale models and improves efficiency and cost-effectiveness. |
Similar Papers
A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Sparse Autoencoders (SAEs) can disentangle complex features into more interpretable components. |
| Approach: | They propose to use Sparse Autoencoders to disentangle LLM features into more interpretable components. |
| Outcome: | The proposed method disentangles complex features into more interpretable components. |
Route Sparse Autoencoder to Interpret Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Sparse autoencoders (SAEs) extract interpretable and monosemantic features in large language models . prior work focused on feature extraction from a single layer, failing to capture activations that span multiple layers. |
| Approach: | They propose a framework that integrates a routing mechanism with a shared SAE to efficiently extract features from multiple layers. |
| Outcome: | The proposed framework extracts features from multiple layers while incurring minimal parameter overhead while achieving high interpretability and flexibility. |
Mechanistic Interpretability Should Prioritize Feature Consistency in Sparse Autoencoders (2026.acl-long)
Copied to clipboard
Xiangchen Song, Aashiq Muhamed, Yujia Zheng, Lingjing Kong, Zeyu Tang, Mona T. Diab, Virginia Smith, Kun Zhang
| Challenge: | Sparse Autoencoders (SAEs) are a tool in mechanistic interpretability (MI) but the aspiration to identify a canonical set of features is challenged by the observed inconsistency of learned SAE features across different training runs. |
| Approach: | They propose to use the Pairwise Dictionary Mean Correlation Coefficient to quantify SAE feature consistency as an evaluation axis alongside reconstruction and sparsity. |
| Outcome: | The proposed measure is based on the pairwise dictionary mean correlation coefficient (PW-MCC) on LLM activations. |
Sparse Autoencoder Features for Classifications and Transferability (2025.emnlp-main)
Copied to clipboard
| Challenge: | Sparse Autoencoders (SAEs) provide potential for uncovering structured, human-interpretable representations in Large Language Models (LLMs). |
| Approach: | They analyze SAEs for interpretable feature extraction from Large Language Models in safety-critical classification tasks. |
| Outcome: | The proposed framework outperforms hidden-state and BoW models while demonstrating cross-lingual toxicity detection and visual classification tasks. |
Unveiling Decision-Making in LLMs for Text Classification : Extraction of influential and interpretable concepts with Sparse Autoencoders (2026.findings-eacl)
Copied to clipboard
| Challenge: | Concept-based explanations for large language models are not well understood in text classification. |
| Approach: | They propose a model with a specialized classifier head and activation rate sparsity loss for sentence classification . they compare it to existing models with HI-Concept and ConceptShap . |
| Outcome: | The proposed model improves both the causality and interpretability of the extracted features. |
Constructing Interpretable Features from Compositional Neuron Groups (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for analyzing LLMs rely on dictionary learning with sparse autoencoders (SAEs) however, SAEs struggle in causal evaluations and lack intrinsic interpretability, as their learning is not explicitly tied to the computations of the model. |
| Approach: | They propose to decompose MLP activations with semi-nonnegative matrix factorization (SNMF) such that the learned features are mapped to their activating inputs, making them directly interpretable. |
| Outcome: | Experiments on Llama 3.1, Gemma 2 and GPT-2 show that SNMF derived features outperform SAEs and a strong supervised baseline on causal steering while aligning with human-interpretable concepts. |
On the Versatility of Sparse Autoencoders for In-Context Learning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Sparse autoencoders (SAEs) are emerging as a key analytical tool in interpretability for large language models. |
| Approach: | They propose to use SAEs to extract knowledge from billions of tokens for sparse reconstruction. |
| Outcome: | The proposed model can extract knowledge from billions of tokens for sparse reconstruction. |
AdaptiveK: Complexity-Driven Sparse Autoencoders for Interpretable Language Model Representations (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to decomposing model activations into interpretable features fail to account for input complexity. |
| Approach: | They propose a framework that dynamically adjusts sparsity levels based on the semantic complexity of each input. |
| Outcome: | The proposed framework outperforms fixed-sparsity approaches on reconstruction fidelity, explained variance, cosine similarity and interpretability metrics while eliminating the burden of extensive hyperparameter tuning. |
Bridging Internal Consistency and External Alignment: A Causal and Dynamic Interpretability Framework for LLM Generation (2026.acl-long)
Copied to clipboard
| Challenge: | Existing interpretability methods focus on internal and external aspects of the model . existing explanations often focus on surface correlations or static dependencies . |
| Approach: | They propose a causal and dynamic interpretability framework for Large Language Models . they characterize backdoor-adjusted causal effects of generated prefix and prompt . |
| Outcome: | The proposed framework provides a unified causal view of internal consistency and external alignment in LLM generation dynamics. |
Measuring What Matters: Evaluating Ensemble LLMs with Label Refinement in Inductive Coding (2025.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are prone to inconsistencies and individual biases, limiting their reliability. |
| Approach: | They propose a framework that combines ensemble methods with code refinement methodology to address these challenges. |
| Outcome: | The proposed framework outperforms large language models and LLMs with a low-rank averaging and a moderator-based mechanism to simulate human consensus. |