The Utility of Interpretability — Emmanuel Amiesen episode artwork

EPISODE · Jun 6, 2025 · 1H 53M

The Utility of Interpretability — Emmanuel Amiesen

from Latent Space: The AI Engineer Podcast · host Latent.Space

Emmanuel Amiesen is lead author of “Circuit Tracing: Revealing Computational Graphs in Language Models” (https://transformer-circuits.pub/2025/attribution-graphs/methods.html ), which is part of a duo of MechInterp papers that Anthropic published in March (alongside https://transformer-circuits.pub/2025/attribution-graphs/biology.html ).We recorded the initial conversation a month ago, but then held off publishing until the open source tooling for the graph generation discussed in this work was released last week: https://www.anthropic.com/research/open-source-circuit-tracingThis is a 2 part episode - an intro covering the open source release, then a deeper dive into the paper — with guest host Vibhu Sapra (https://x.com/vibhuuuus ) and Mochi the MechInterp Pomsky (https://x.com/mochipomsky ). Thanks to Vibhu for making this episode happen!While the original blogpost contained some fantastic guided visualizations (which we discuss at the end of this pod!), with the notebook and Neuronpedia visualization (https://www.neuronpedia.org/gemma-2-2b/graph ) released this week, you can now explore on your own with Neuronpedia, as we show you in the video version of this pod.Full Video EpisodeTimestamps00:00 Intro & Guest Introductions01:00 Anthropic's Circuit Tracing Release06:11 Exploring Circuit Tracing Tools & Demos13:01 Model Behaviors and User Experiments17:02 Behind the Research: Team and Community24:19 Main Episode Start: Mech Interp Backgrounds25:56 Getting Into Mech Interp Research31:52 History and Foundations of Mech Interp37:05 Core Concepts: Superposition & Features39:54 Applications & Interventions in Models45:59 Challenges & Open Questions in Interpretability57:15 Understanding Model Mechanisms: Circuits & Reasoning01:04:24 Model Planning, Reasoning, and Attribution Graphs01:30:52 Faithfulness, Deception, and Parallel Circuits01:40:16 Publishing Risks, Open Research, and Visualization01:49:33 Barriers, Vision, and Call to Action This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.latent.space/subscribe

Episode metadata supplied by the publisher feed · Published Jun 6, 2025

Embed this episode

NOW PLAYING

The Utility of Interpretability — Emmanuel Amiesen

0:00 1:53:02

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 Latent Space: The AI Engineer Podcast?

This episode is 1 hour and 53 minutes long.

When was this Latent Space: The AI Engineer Podcast episode published?

This episode was published on June 6, 2025.

Can I download this Latent Space: The AI Engineer Podcast episode?

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