Evaluating Synthetic Activations composed of SAE Latents in GPT-2

23 September 2024

Giorgi Giglemiani

Nora Petrova

Chatrik Singh Mangat

Jett Janiak

Stefan Heimersheim

Papers citing "Evaluating Synthetic Activations composed of SAE Latents in GPT-2"

2 / 2 papers shown

Title
Decomposing The Dark Matter of Sparse Autoencoders Joshua Engels Logan Riggs Max Tegmark LLMSV 57 9 0 18 Oct 2024
Investigating Sensitive Directions in GPT-2: An Improved Baseline and Comparative Analysis of SAEs Daniel J. Lee Stefan Heimersheim AAML 24 4 0 16 Oct 2024