sparse autoencoder

Understanding Emergent Misalignment via Feature Superposition Geometry ACL 2026 Unveiling Decision-Making in LLMs for Text Classification : Extraction of influential and interpretable concepts with Sparse Autoencoders EACL 2026 Unveiling Intrinsic Dimension of Texts: from Academic Abstract to Creative Story EACL 2026 Analyze Feature Flow to Enhance Interpretation and Steering in Language Models ICML 2025 Analyzing (In)Abilities of SAEs via Formal Languages NAACL 2025 Are Sparse Autoencoders Useful? A Case Study in Sparse Probing ICML 2025 Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective ACL 2025 A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models EMNLP 2025 Automatically Interpreting Millions of Features in Large Language Models ICML 2025 AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders ICML 2025 Beyond Input Activations: Identifying Influential Latents by Gradient Sparse Autoencoders EMNLP 2025 Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms ACL 2025 Breaking Bad Tokens: Detoxification of LLMs Using Sparse Autoencoders EMNLP 2025 CE-Bench: Towards a Reliable Contrastive Evaluation Benchmark of Interpretability of Sparse Autoencoders EMNLP 2025 Compute Optimal Inference and Provable Amortisation Gap in Sparse Autoencoders ICML 2025 Constrain Alignment with Sparse Autoencoders ICML 2025 Decoding Dark Matter: Specialized Sparse Autoencoders for Interpreting Rare Concepts in Foundation Models NAACL 2025 Decoding Dense Embeddings: Sparse Autoencoders for Interpreting and Discretizing Dense Retrieval EMNLP 2025 Disentangling language change: sparse autoencoders quantify the semantic evolution of indigeneity in French NAACL 2025 Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models ICLR 2025 Efficient Dictionary Learning with Switch Sparse Autoencoders ICLR 2025 Extended Abstract for “Linguistic Universals”: Emergent Shared Features in Independent Monolingual Language Models via Sparse Autoencoders EMNLP 2025 FADE: Why Bad Descriptions Happen to Good Features ACL 2025 FaithfulSAE: Towards Capturing Faithful Features with Sparse Autoencoders without External Datasets Dependency ACL 2025 Feature Extraction and Steering for Enhanced Chain-of-Thought Reasoning in Language Models EMNLP 2025