Discover this podcast and so much more

Podcasts are free to enjoy without a subscription. We also offer ebooks, audiobooks, and so much more for just $11.99/month.

E48: Mechanizing Mechanistic Interpretability with Arthur Conmy

E48: Mechanizing Mechanistic Interpretability with Arthur Conmy

From"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis


E48: Mechanizing Mechanistic Interpretability with Arthur Conmy

From"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

ratings:
Length:
134 minutes
Released:
Jul 27, 2023
Format:
Podcast episode

Description

Arthur Conmy sits down with Nathan Labenz for an accessible deep dive into the state of interpretability research online today. They discuss how pioneering researchers have painstakingly worked to isolate the sub-circuits within transformers that are responsible for different aspects of AI capabilities. Arthur also introduces us to a new ACDC approach that he and his co-authors have taken to automating some of the most time-consuming parts of this work. If you’re looking for an ERP platform, check out our sponsor, NetSuite: http://netsuite.com/cognitive

The Cognitive Revolution is a part of the Turpentine podcast network. Learn more: Turpentine.co

TIMESTAMPS:
(00:00) Episode Preview
(04:40) What attracted Arthur to mechanistic interpretability?
(07:49) LLM information processing: General Understanding vs Stochastic Parrot Paradigm
(14:00) ACDC paper: https://arxiv.org/abs/2304.14997 
(14:45) Sponsors: NetSuite | Omneky
(24:30) Putting together data sets 
(32:39) How to intervene in LLMs network activity
(36:00) Setting metrics to evaluate the production of correct completions
(44:20) The future of the mechanistic interpretability research 
(50:00) Extracting upstream activations in the ACDC project and evaluating impact on downstream components.
(56:00) Anthropic research findings 
(01:08:00) 3-Step process of the ACDC approach
(01:22:00) Setting a threshold and validation
(01:27:00) Goal of the approach
(01:32:00) Compute requirements
(01:35:30) Scaling laws for mechanistic interpretability
(01:40:00) Accessibility of this research for casual enthusiasts
(01:46:00) Emergence discourse
(01:56:00) Path to AI safety

LINKS:
https://arthurconmy.github.io/

X:
@labenz (Nathan)
@arthurconmy (Arthur)
@cogrev_podcast

SPONSORS: NetSuite | Omneky
-NetSuite provides financial software for all your business needs. More than thirty-six thousand companies have already upgraded to NetSuite, gaining visibility and control over their financials, inventory, HR, eCommerce, and more. If you’re looking for an ERP platform: NetSuite (Code http://netsuite.com/cognitive) and defer payments of a FULL NetSuite implementation for six months.

-Omneky is an omnichannel creative generation platform that lets you launch hundreds of thousands of ad iterations that *actually work* customized across all platforms, with a click of a button. Omneky combines generative AI and real-time advertising data. Mention "Cog Rev" for 10% off.
Released:
Jul 27, 2023
Format:
Podcast episode

Titles in the series (100)

A weekly podcast where hosts Erik Torenberg and Nathan Labenz interview the builders on the edge of AI and explore the dramatic shift it will unlock in the coming years.