Search Results for author:
Found 6 papers, 1 papers with code
SAEBench: A Comprehensive Benchmark for Sparse Autoencoders in Language Model Interpretability
We introduce SAEBench, a comprehensive evaluation suite that measures SAE performance across seven diverse metrics, spanning interpretability, feature disentanglement and practical applications like unlearning.
Sparse Autoencoders Do Not Find Canonical Units of Analysis
Using meta-SAEs -- SAEs trained on the decoder matrix of another SAE -- we find that latents in SAEs often decompose into combinations of latents from a smaller SAE, showing that larger SAE latents are not atomic.
LLM Circuit Analyses Are Consistent Across Training and Scale
These results suggest that circuit analyses conducted on small models at the end of pre-training can provide insights that still apply after additional pre-training and over model scale.
Transformer-Based Models Are Not Yet Perfect At Learning to Emulate Structural Recursion
The framework includes a representation that captures the general \textit{syntax} of structural recursion, coupled with two different frameworks for understanding their \textit{semantics} -- one that is more natural from a programming languages perspective and one that helps bridge that perspective with a mechanistic understanding of the underlying transformer architecture.
Linear Representations of Sentiment in Large Language Models
Sentiment is a pervasive feature in natural language text, yet it is an open question how sentiment is represented within Large Language Models (LLMs).
Can Transformers Learn to Solve Problems Recursively?
Neural networks have in recent years shown promise for helping software engineers write programs and even formally verify them.
