"Sparse Autoencoders Find Highly Interpretable Directions in Language Models" by Logan Riggs et al episode artwork

EPISODE · Sep 27, 2023 · 10 MIN

"Sparse Autoencoders Find Highly Interpretable Directions in Language Models" by Logan Riggs et al

from LessWrong (Curated & Popular) · host LessWrong

This is a linkpost for Sparse Autoencoders Find Highly Interpretable Directions in Language ModelsWe use a scalable and unsupervised method called Sparse Autoencoders to find interpretable, monosemantic features in real LLMs (Pythia-70M/410M) for both residual stream and MLPs. We showcase monosemantic features, feature replacement for Indirect Object Identification (IOI), and use OpenAI's automatic interpretation protocol to demonstrate a significant improvement in interpretability.Source:https://www.lesswrong.com/posts/Qryk6FqjtZk9FHHJR/sparse-autoencoders-find-highly-interpretable-directions-inNarrated for LessWrong by TYPE III AUDIO.Share feedback on this narration.[125+ Karma Post] ✓

Episode metadata supplied by the publisher feed · Published Sep 27, 2023

Embed this episode

This is a linkpost for Sparse Autoencoders Find Highly Interpretable Directions in Language Models We use a scalable and unsupervised method called Sparse Autoencoders to find interpretable, monosemantic features in real LLMs (Pythia-70M/410M) for both residual stream and MLPs. We showcase monosemantic features, feature replacement for Indirect Object Identification (IOI), and use OpenAI's automatic interpretation protocol to demonstrate a significant improvement in interpretability. Source:...

Distinct summary based on available episode metadata or transcript content.

NOW PLAYING

"Sparse Autoencoders Find Highly Interpretable Directions in Language Models" by Logan Riggs et al

0:00 10:12

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

No similar podcasts found.

Frequently Asked Questions

How long is this episode of LessWrong (Curated & Popular)?

This episode is 10 minutes long.

When was this LessWrong (Curated & Popular) episode published?

This episode was published on September 27, 2023.

Can I download this LessWrong (Curated & Popular) episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!