Jordan Taylor, Sid Black, Dillon Bowen, Thomas Read, Satvik Golechha, Alex Zelenka-Martin, Oliver Makins, Connor Kissane, Kola Ayonrinde, Jacob Merizian, Samuel Marks, Chris Cundy, Joseph Bloom UK AI Security Institute, FAR.AI, Anthropic Links: Paper | Code | Models | Transcripts | Interactive Demo Epistemic Status: We're sharing our paper and...
Introduction Joseph Bloom, Alan Cooney This is a research update from the White Box Control team at UK AISI. In this update, we share preliminary results on the topic of sandbagging that may be of interest to researchers working in the field. The format of this post was inspired by...
A short summary of the paper is presented below. This work was produced by Apollo Research in collaboration with Jordan Taylor (MATS + University of Queensland) . TL;DR: We propose end-to-end (e2e) sparse dictionary learning, a method for training SAEs that ensures the features learned are functionally important by minimizing...
Jordan Taylor, Sid Black, Dillon Bowen, Thomas Read, Satvik Golechha, Alex Zelenka-Martin, Oliver Makins, Connor Kissane, Kola Ayonrinde, Jacob Merizian, Samuel Marks, Chris Cundy, Joseph Bloom
UK AI Security Institute, FAR.AI, Anthropic
Links: Paper | Code | Models | Transcripts | Interactive Demo
Epistemic Status: We're sharing our paper and a hastily written summary of it, assuming a higher level of context on sandbagging / auditing games than other materials. We also share some informal commentary on our results. This post was written by Jordan and Joseph, and may not reflect the views of all authors.
This summary diverges from the paper, condensing heavily while adding our additional commentary. The limitations and future work sections are... (read 2963 more words →)
Joseph Bloom, Alan Cooney
This is a research update from the White Box Control team at UK AISI. In this update, we share preliminary results on the topic of sandbagging that may be of interest to researchers working in the field.
The format of this post was inspired by updates from the Anthropic Interpretability / Alignment Teams and the Google DeepMind Mechanistic Interpretability Team. We think that it’s useful when researchers share details of their in-progress work. Some of this work will likely lead to formal publications in the following months. Please interpret these results as you might a colleague sharing their lab notes.
As this is our first such progress update, we also include some paragraphs introducing the team and contextualising our work.
A short summary of the paper is presented below.
This work was produced by Apollo Research in collaboration with Jordan Taylor (MATS + University of Queensland) .
TL;DR: We propose end-to-end (e2e) sparse dictionary learning, a method for training SAEs that ensures the features learned are functionally important by minimizing the KL divergence between the output distributions of the original model and the model with SAE activations inserted. Compared to standard SAEs, e2e SAEs offer a Pareto improvement: They explain more network performance, require fewer total features, and require fewer simultaneously active features per datapoint, all with no cost to interpretability. We explore geometric and qualitative differences between e2e SAE features and standard SAE... (read 1115 more words →)