Many distinct SAE features share identical explanations, with the average annotation resolving only 70% of feature identity in a large annotated dataset.
arXiv:2410.20526 (2024)
18 Pith papers cite this work, alongside 3 external citations. Polarity classification is still indexing.
18
Pith papers citing it
3
external citations · external index
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Auto-interpretation labels for SAE features generalize poorly across languages and scripts, missing the same semantic content up to 4x more often in Serbian than English and more in Cyrillic than Latin despite deterministic transliteration.
SAEs used for layer selection with raw task vectors outperform subspace projection and raise math reasoning accuracy on Gemma-3-4B-IT.
ReSAEs improve multi-layer SAE interventions on Pythia-1.4B and Gemma-2-9B by training later-layer dictionaries on residuals after affine mapping, recovering more cross-entropy loss despite lower raw variance reconstruction.
LLMs process negation using both attention-based suppression and constructive representation mechanisms (construction dominant), with late-layer attention shortcuts explaining poor accuracy on negation tasks.
A four-step recipe partitions the input space using interchange intervention behavior to diagnose where causal abstractions hold and to guide improvements, demonstrated by recovering a full hypothesis from scratch in a toy logic task.
Uncertainty and correctness in LLMs are encoded by distinct feature populations, with suppression of confounded features improving accuracy and reducing entropy.
Multilingual SAEs strengthen cross-lingual representations for reliable steering and an intersection-based rule selects effective layers without exhaustive search.
A steering-grid protocol that varies coefficient and joint feature set shows that standard single-feature SAE labels are often incomplete, naming an activation regime rather than a causal axis.
Distinct linear knowledge vectors for deductive, inductive, and abductive reasoning in LLMs can be refined via complementary subspace constraints to improve performance through mutual knowledge sharing.
MEDS improves LLM RL performance by up to 4.13 pass@1 and 4.37 pass@128 points by dynamically penalizing rollouts matching prevalent historical error clusters identified via memory-stored representations and density clustering.
NeuroCogMap maps LLM internal representations into stable functional parcels tied to cognitive functions, failure modes, and human cortical activity during language tasks.
The survey organizes mechanistic interpretability techniques into a Locate-Steer-Improve framework to enable actionable improvements in LLM alignment, capability, and efficiency.
Safe-SAIL supplies a pre-explanation metric and segment-level simulation to interpret 1758 safety SAE features across pornography, politics, violence, and terror, with public models and tools released.
An autonomy-qualified Second Welfare Theorem is stated for post-AGI economies under the joint conditions of convexity, stable moral status, non-fungible rights, welfare selection, non-manipulation, governed self-modification, and verification.
Steering is positioned as a distinct adaptation paradigm that uses targeted activation interventions for local, reversible behavioral changes without parameter updates.
Trains and releases SAEs for Qwen3-1.7B/4B/8B models with layer-wise coverage and demonstrates causal steering of refusal via selected features.
citing papers explorer
-
Descriptive Collision in Sparse Autoencoder Auto-Interpretability: When One Explanation Describes Many Features
Many distinct SAE features share identical explanations, with the average annotation resolving only 70% of feature identity in a large annotated dataset.
-
How Far Do Auto-Interpretation Labels Generalize: A Controlled Study Across Languages, Scripts, and Rewordings
Auto-interpretation labels for SAE features generalize poorly across languages and scripts, missing the same semantic content up to 4x more often in Serbian than English and more in Cyrillic than Latin despite deterministic transliteration.
-
Interpretability-Guided Layer Selection over Subspace Projection: SAEs as Stethoscopes, Not Scalpels, for Raw Task Vector Model Editing
SAEs used for layer selection with raw task vectors outperform subspace projection and raise math reasoning accuracy on Gemma-3-4B-IT.
-
ReSAE: Residualized Sparse Autoencoders for Multi-Layer Transformer Interventions
ReSAEs improve multi-layer SAE interventions on Pythia-1.4B and Gemma-2-9B by training later-layer dictionaries on residuals after affine mapping, recovering more cross-entropy loss despite lower raw variance reconstruction.
-
How Language Models Process Negation
LLMs process negation using both attention-based suppression and constructive representation mechanisms (construction dominant), with late-layer attention shortcuts explaining poor accuracy on negation tasks.
-
Bucketing the Good Apples: A Method for Diagnosing and Improving Causal Abstraction
A four-step recipe partitions the input space using interchange intervention behavior to diagnose where causal abstractions hold and to guide improvements, demonstrated by recovering a full hypothesis from scratch in a toy logic task.
-
Are LLM Uncertainty and Correctness Encoded by the Same Features? A Functional Dissociation via Sparse Autoencoders
Uncertainty and correctness in LLMs are encoded by distinct feature populations, with suppression of confounded features improving accuracy and reducing entropy.
-
Multilingual Steering by Design: Multilingual Sparse Autoencoders and Principled Layer Selection
Multilingual SAEs strengthen cross-lingual representations for reliable steering and an intersection-based rule selects effective layers without exhaustive search.
-
Steering grids for sparse-autoencoder features: when a top-context label names an activation regime rather than a causal axis
A steering-grid protocol that varies coefficient and joint feature set shows that standard single-feature SAE labels are often incomplete, naming an activation regime rather than a causal axis.
-
Knowledge Vector of Logical Reasoning in Large Language Models
Distinct linear knowledge vectors for deductive, inductive, and abductive reasoning in LLMs can be refined via complementary subspace constraints to improve performance through mutual knowledge sharing.
-
The Past Is Not Past: Memory-Enhanced Dynamic Reward Shaping
MEDS improves LLM RL performance by up to 4.13 pass@1 and 4.37 pass@128 points by dynamically penalizing rollouts matching prevalent historical error clusters identified via memory-stored representations and density clustering.
-
NeuroCogMap Reveals Cognitive Organization of Large Language Models
NeuroCogMap maps LLM internal representations into stable functional parcels tied to cognitive functions, failure modes, and human cortical activity during language tasks.
-
Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models
The survey organizes mechanistic interpretability techniques into a Locate-Steer-Improve framework to enable actionable improvements in LLM alignment, capability, and efficiency.
-
Safe-SAIL: Towards a Fine-grained Safety Landscape of Large Language Models via Sparse Autoencoder Interpretation Framework
Safe-SAIL supplies a pre-explanation metric and segment-level simulation to interpret 1758 safety SAE features across pornography, politics, violence, and terror, with public models and tools released.
-
Post-AGI Economies: Superposition and the Second Fundamental Theorem of Welfare Economics
An autonomy-qualified Second Welfare Theorem is stated for post-AGI economies under the joint conditions of convexity, stable moral status, non-fungible rights, welfare selection, non-manipulation, governed self-modification, and verification.
-
From Weights to Activations: Is Steering the Next Frontier of Adaptation?
Steering is positioned as a distinct adaptation paradigm that uses targeted activation interventions for local, reversible behavioral changes without parameter updates.
-
Discovering Millions of Interpretable Features with Sparse Autoencoders
Trains and releases SAEs for Qwen3-1.7B/4B/8B models with layer-wise coverage and demonstrates causal steering of refusal via selected features.
- Sign-Aware Gated Sparse Autoencoders: Modeling Anticorrelated Features with Bi-Jump-ReLU Activations