SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.06453 · 2024

PURE: Turning Polysemantic Neurons Into Pure Features by Identifying Relevant Circuits

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 4 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
maxdreyer/pure canonical 3 of 4
FunctionStatusWhere it lives
compute_distances Ran maxdreyer/pure/experiments/disentangling/eval_CLIP_alignment.py
code served (permissive licence) · get_code("3879cca40ffb1d1e")
load_config Ran maxdreyer/pure/utils/helper.py
code served (permissive licence) · get_code("46ba6f801029f1b7")
mystroke Ran maxdreyer/pure/utils/render.py
code served (permissive licence) · get_code("a308f46e6162ffd2")
get_imagenet Not yet run maxdreyer/pure/datasets/imagenet.py
code served (permissive licence) · get_code("dfa5a65342a498e4")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The field of mechanistic interpretability aims to study the role of individual neurons in Deep Neural Networks. Single neurons, however, have the capability to act polysemantically and encode for multiple (unrelated) features, which renders their interpretation difficult. We present a method for disentangling polysemanticity of any Deep Neural Network by decomposing a polysemantic neuron into multiple monosemantic "virtual" neurons. This is achieved by identifying the relevant sub-graph ("circuit") for each "pure" feature. We demonstrate how our approach allows us to find and disentangle various polysemantic units of ResNet models trained on ImageNet. While evaluating feature visualizations using CLIP, our method effectively disentangles representations, improving upon methods based on neuron activations. Our code is available at https://github.com/maxdreyer/PURE.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2404.06453")
get_code_for_paper("2404.06453")
have("2404.06453")

Connect an agent — have() is free.