World Model · podcast knowledge graph

Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety

2024-08-17 · 112 min · 19 entities

episode page audio

Asserted relationships

  • → hosted by Erik Torenberg, Nathan Labenz person
    0.55
    evidence rules-v4
    Feed author/publisher: Erik Torenberg, Nathan Labenz
  • → discusses Society & Culture concept
    0.40
    evidence rules-v4
    Feed category: Society & Culture
  • → discusses Technology concept
    0.40
    evidence rules-v4
    Feed category: Technology
  • → discusses Business concept
    0.40
    evidence rules-v4
    Feed category: Business
  • → discusses Entrepreneurship concept
    0.40
    evidence rules-v4
    Feed category: Entrepreneurship
  • → hosted by Turpentine company
    0.40
    evidence rules-v4
    Feed author/publisher: Turpentine
  • → references omneky.com website
    0.38
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://omneky.com/
  • → references choosesquad.com website
    0.38
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://choosesquad.com/
  • → references JCkphVqj website
    0.38
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://hmplogxqz0y.typeform.com/to/JCkphVqj

Entities found in this episode

concepts 9

  • mentioned Society & Culture concept
    0.50
    evidence rules-v4
    Feed category: Society & Culture
  • mentioned Technology concept
    0.50
    evidence rules-v4
    Feed category: Technology
  • mentioned Business concept
    0.50
    evidence rules-v4
    Feed category: Business
  • mentioned Entrepreneurship concept
    0.50
    evidence rules-v4
    Feed category: Entrepreneurship
  • 0.42
    evidence rules-v4
    Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety
  • discusses Society & Culture concept
    0.40
    evidence rules-v4
    Feed category: Society & Culture
  • discusses Technology concept
    0.40
    evidence rules-v4
    Feed category: Technology
  • discusses Business concept
    0.40
    evidence rules-v4
    Feed category: Business
  • discusses Entrepreneurship concept
    0.40
    evidence rules-v4
    Feed category: Entrepreneurship

websites 6

  • mentioned omneky.com website
    0.45
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://omneky.com/
  • mentioned choosesquad.com website
    0.45
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://choosesquad.com/
  • mentioned JCkphVqj website
    0.45
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://hmplogxqz0y.typeform.com/to/JCkphVqj
  • references omneky.com website
    0.38
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://omneky.com/
  • references choosesquad.com website
    0.38
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://choosesquad.com/
  • references JCkphVqj website
    0.38
    evidence rules-v4
    Link in episode "Popular Mechanistic Interpretability: Goodfire Lights the Way to AI Safety": https://hmplogxqz0y.typeform.com/to/JCkphVqj

persons 2

companys 2

  • mentioned Turpentine company
    0.70
    evidence rules-v4
    Feed author/publisher: Turpentine
  • hosted by Turpentine company
    0.40
    evidence rules-v4
    Feed author/publisher: Turpentine
Episode description as stored
Nathan explores the cutting-edge field of mechanistic interpretability with Dan Balsam and Tom McGrath, co-founders of Goodfire. In this episode of The Cognitive Revolution, we delve into the science of understanding AI models' inner workings, recent breakthroughs, and the potential impact on AI safety and control. Join us for an insightful discussion on sparse autoencoders, polysemanticity, and the future of interpretable AI. Papers Very accessible article on types of representations: Local vs Distributed Coding Theoretical understanding of how models might pack concepts into their representations: Toy Models of Superposition How structure in the world gives rise to structure in the latent space: The Geometry of Categorical and Hierarchical Concepts in Large Language Models Using sparse autoencoders to pull apart language model representations: Sparse Autoencoders / Towards Monosemanticity / Scaling Monosemanticity Finding & teaching concepts in superhuman systems: Acquisition of Chess Knowledge in AlphaZero / Bridging the Human-AI Knowledge Gap: Concept Discovery and Transfer in AlphaZero Connecting microscopic learning to macroscopic phenomena: The Quantization Model of Neural Scaling Understanding at scale: Language models can explain neurons in language models Apply to join over 400 founders and execs in the Turpentine Network: https://hmplogxqz0y.typeform.com/to/JCkphVqj SPONSORS: Oracle Cloud Infrastructure (OCI) is a single platform for your infrastructure, database, application development, and AI needs. OCI has four to eight times the bandwidth of other clouds; offers one consistent price, and nobody does data better than Oracle. If you want to do more and spend less, take a free test drive of OCI at https://oracle.com/cognitive The Brave search API can be used to assemble a data set to train your AI models and help with retrieval augmentation at the time of inference. All while remaining affordable with developer first pricing, integrating the Brave search API into your workflow translates to more ethical data sourcing and more human representative data sets. Try the Brave search API for free for up to 2000 queries per month at https://bit.ly/BraveTCR 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 https://www.omneky.com/ Head to Squad to access global engineering without the headache and at a fraction of the cost: head to https://choosesquad.com/ and mention “Turpentine” to skip the waitlist. CHAPTERS: (00:00:00) About the Show (00:00:22) About the Episode (00:03:52) Introduction and Background (00:08:43) State of Interpretability Research (00:12:06) Key Insights in Interpretability (00:16:53) Polysemanticity and Model Compression (Part 1) (00:17:00) Sponsors: Oracle | Brave (00:19:04) Polysemanticity and Model Compression (Part 2) (00:22:50) Sparse Autoencoders Explained (00:27:19) Challenges in Interpretability Research (Part 1) (00:30:54) Sponsors: Omneky | Squad (00:32:41) Challenges in Interpretability Research (Part 2) (00:33:51) Goodfire's Vision and Mission (00:37:08) Interpretability and Scientific Models (00:43:48) Architecture and Interpretability Techniques (00:50:08) Quantization and Model Representation (00:54:07) Future of Interpretability Research (01:01:38) Skepticism and Challenges in Interpretability (01:07:51) Alternative Architectures and Universality (01:13:39) Goodfire's Business Model and Funding (01:18:47) Building the Team and Future Plans (01:31:03) Hiring and Getting Involved in Interpretability (01:51:28) Closing Remarks (01:51:38) Outro