sparse autoencoder

Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders ACL 2025 Features that Make a Difference: Leveraging Gradients for Improved Dictionary Learning NAACL 2025 From Mechanistic Interpretability to Mechanistic Biology: Training, Evaluating, and Interpreting Sparse Autoencoders on Protein Language Models ICML 2025 Group-SAE: Efficient Training of Sparse Autoencoders for Large Language Models via Layer Groups EMNLP 2025 How a Bilingual LM Becomes Bilingual: Tracing Internal Representations with Sparse Autoencoders EMNLP 2025 Improving LLM Reasoning through Interpretable Role-Playing Steering EMNLP 2025 Incremental Sentence Processing Mechanisms in Autoregressive Transformer Language Models NAACL 2025 Inference-Time Decomposition of Activations (ITDA): A Scalable Approach to Interpreting Large Language Models ICML 2025 Interpretable Company Similarity with Sparse Autoencoders ACL 2025 Interpretable Sparse Features for Probing Self-Supervised Speech Models IJCNLP 2025 Interpret and Control Dense Retrieval with Sparse Latent Features NAACL 2025 Interpreting CLIP with Hierarchical Sparse Autoencoders ICML 2025 Interpreting Language Models Through Concept Descriptions: A Survey EMNLP 2025 Large Language Models Share Representations of Latent Grammatical Concepts Across Typologically Diverse Languages NAACL 2025 Large Multi-modal Models Can Interpret Features in Large Multi-modal Models ICCV 2025 Latent Inter-User Difference Modeling for LLM Personalization EMNLP 2025 Learning Multi-Level Features with Matryoshka Sparse Autoencoders ICML 2025 LinguaLens: Towards Interpreting Linguistic Mechanisms of Large Language Models via Sparse Auto-Encoder EMNLP 2025 Low-Rank Adapting Models for Sparse Autoencoders ICML 2025 Mechanistic Permutability: Match Features Across Layers ICLR 2025 MIB: A Mechanistic Interpretability Benchmark ICML 2025 MicroEdit: Neuron-level Knowledge Disentanglement and Localization in Lifelong Model Editing EMNLP 2025 Model Unlearning via Sparse Autoencoder Subspace Guided Projections EMNLP 2025 Not All Language Model Features Are One-Dimensionally Linear ICLR 2025 On the Versatility of Sparse Autoencoders for In-Context Learning EMNLP 2025