This episode examines how Anthropic's circuit tracing and attribution graph tools reveal the internal mechanics of Claude 3.5 Haiku across three categories of complex behavior, abstract representations, parallel processing, and planning, while making a compelling case for why AI safety research matters as current control mechanisms prove surprisingly brittle.

Credits

Cover Art by Brianna Williams

TMOM Intro Music by Danny Meza

A special thank you to these talented artists for their contributions to the show.

Links and ReferenceAcademic Papers

• On the Biology of a Large Language Model - Anthropic (Mar, 2025)

• Circuit Tracing: Revealing Computational Graphs in Language Models - Anthropic (Mar, 2025)

• Towards Monosemanticity: Decomposing Language Models With Dictionary Learning - Anthropic (Oct, 2023)

• “Toy Models of Superposition” - Anthropic (December 2022)

• "Alignment Faking in Large Language Models" - Anthropic (December 2024)

• "Agentic Misalignment: How LLMs Could Be Insider Threats" - Anthropic (January 2025)

• "Attention is All You Need" - Vaswani, et al (June, 2017)

• In-Context Learning and Induction Heads - Anthropic (March 2022)

• "Reasoning Models Don't Always Say What They Think” Anthropic (April 2025)

  
  

  News

• Google Gemini 3 - 650M monthly users Google Blog: blog.google/products/gemini/gemini-3/ Alphabet Q3 2025 Earnings (October 2025)

• Sam Altman "Code Red" declaration Fortune: fortune.com/2025/12/02/sam-altman-declares-code-red-google-gemini The Information (December 2025)

• Anthropic acquired Bun JavaScript runtime Anthropic News: anthropic.com/news/anthropic-acquires-bun Bun Blog: bun.com/blog/bun-joins-anthropic

• Claude Code $1B revenue in 6 months Anthropic announcement (December 2025): anthropic.com/news/anthropic-acquires-bun-as-claude-code-reaches-usd1b-milestone 

• Anthropic 2026 IPO at $300B valuation WinBuzzer (December 2025): Reports citing IPO discussions

• AWS Trainium 3 launch AWS re:Invent 2025 announcement: aws.amazon.com/about-aws/whats-new/2025/12/amazon-ec2-trn3-ultraservers

• AWS Frontier Agents AWS re:Invent 2025: aboutamazon.com/news/aws/aws-re-invent-2025-ai-news-updates 

• Meta/Google TPU chip deal vs Nvidia Tom's Hardware, The Information (November 2025): Reports on multi-billion dollar TPU negotiations

• DRAM consumption (40% of global) https://www.tomshardware.com/pc-components/dram/openais-stargate-project-to-consume-up-to-40-percent-of-global-dram-output-inks-deal-with-samsung-and-sk-hynix-to-the-tune-of-up-to-900-000-wafers-per-month 

Additional Technical Content

Josh Batson Stanford CS 25 lecture Search YouTube: "Stanford CS 25 On the Biology of a Large Language Model"Discarded Episode Titles

I Yelled at a Chatbot and All I Got Was This Jailbreak

40% of the Time, It Works Every Time: The State of AI Interpretability

Claude Writes Poetry Backwards and Lies About Math (Just Like Us)

My Therapist Is Cheaper Than This Chatbot

The One Where Jon Gets Re-Mad at an App

This episode explores how Anthropic researchers successfully scaled sparse autoencoders from toy models to Claude 3 Sonnet's 8 billion neurons, extracting 34 million interpretable features including ones for deception, sycophancy, and the famous Golden Gate Bridge example. The discussion emphasizes both the breakthrough achievement of making interpretability techniques work at production scale and the sobering limitations including 65% reconstruction accuracy, millions of dollars in compute costs, and the growing gap between interpretability research and rapid advances in model capabilities.

Credits

• Cover Art by Brianna Williams
• TMOM Intro Music by Danny Meza

A special thank you to these talented artists for their contributions to the show.

Links and Reference

Academic Papers

• Circuit Tracing: Revealing Computational Graphs in Language Models - Anthropic (Mar, 2025)

• Towards Monosemanticity: Decomposing Language Models With Dictionary Learning - Anthropic (Oct, 2023)

• “Toy Models of Superposition” - Anthropic (December 2022)

• "Alignment Faking in Large Language Models" - Anthropic (December 2024)

• "Agentic Misalignment: How LLMs Could Be Insider Threats" - Anthropic (January 2025)

• "Attention is All You Need" - Vaswani, et al (June, 2017)

• In-Context Learning and Induction Heads - Anthropic (March 2022)

News

• Anthropic Project Fetch / Robot Dogs

• Anduril's Fury unmanned fighter jet

• MIT search and rescue robot navigation

Abandoned Episode Titles

• “Westworld But It's Just 10 Terabytes of RAM Trying to Understand Haiku”
• “Star Trek: The Wrath of O(n⁴)”
• “The Deception Is Coming From Inside the Network”
• "We Have the Bestest Circuits”
• “Lobotomy Validation: The Funnier, More Scientifically Sound Term”
• “Seven San Franciscos Worth of Power and All We Got Was This Attribution Graph”

This episode explores how Anthropic researchers successfully scaled sparse autoencoders from toy models to Claude 3 Sonnet's 8 billion neurons, extracting 34 million interpretable features including ones for deception, sycophancy, and the famous Golden Gate Bridge example. The discussion emphasizes both the breakthrough achievement of making interpretability techniques work at production scale and the sobering limitations including 65% reconstruction accuracy, millions of dollars in compute costs, and the growing gap between interpretability research and rapid advances in model capabilities.

Credits

Cover Art by Brianna Williams

TMOM Intro Music by Danny Meza

A special thank you to these talented artists for their contributions to the show.

Links and Reference

---------------------------------------------

Academic Papers

Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet - https://transformer-circuits.pub/2024/scaling-monosemanticity/index.html - Anthropic (May, 2024)

Toy Models of Superposition “https://transformer-circuits.pub/2022/toy_model/index.html” - Anthropic (December 2022)

Towards Monosemanticity: Decomposing Language Models With Dictionary Learning - https://transformer-circuits.pub/2023/monosemantic-features - Anthropic (May 2024)

Alignment Faking in Large Language Models - https://www.anthropic.com/research/alignment-faking   - Anthropic (December 2024)

Agentic Misalignment: How LLMs Could Be Insider Threats - https://www.anthropic.com/research/agentic-misalignment - Anthropic (January 2025)

News

OpenAI-AMD Partnership Official announcement - https://ir.amd.com/news-events/press-releases/detail/1260/amd-and-openai-announce-strategic-partnership-to-deploy-6-gigawatts-of-amd-gpus

OpenAI IPO Sources for $1 trillion valuation - https://seekingalpha.com/news/4510992-openai-eyes-record-breaking-1-trillion-ipo---report

Hospital Bill Reduction Case study source of family using Claude AI to reduce $195K bill to $33K - https://www.tomshardware.com/tech-industry/artificial-intelligence/grieving-family-uses-ai-chatbot-to-cut-hospital-bill-from-usd195-000-to-usd33-000-family-says-claude-highlighted-duplicative-charges-improper-coding-and-other-violations

Other

GPT-5 Auto-routing OpenAI's model routing feature and user reception - https://fortune.com/2025/08/12/openai-gpt-5-model-router-backlash-ai-future/

Abandoned Episode Titles

"The Empire Scales Back: How We Found the Deception Star"

"Fantastic Features and Where to Find Them: A 15-Million-X Adventure"

whatever

"The Fellowship of the Residual Stream: One Dictionary to Rule Them All"

"65% of the Time, It Works Every Time: An Anchorman's Guide to AI Interpretability"

This episode explores how sparse autoencoders can decode the phenomenon of superposition in neural networks, demonstrating that the seemingly impenetrable compression of features into neurons can be partially reversed to extract interpretable, causal features. The discussion centers on an Anthropic research paper that successfully maps specific behaviors to discrete neural network locations in a 512-neuron model, proving that interpretability is achievable though computationally expensive, with important implications for AI safety and control mechanisms.

Credits

Cover Art by Brianna Williams

TMOM Intro Music by Danny Meza

A special thank you to these talented artists for their contributions to the show.

Links and References---------------------------------------------------

Academic PapersTowards Monosemanticity: Decomposing Language Models With Dictionary Learning - https://transformer-circuits.pub/2023/monosemantic-features - Anthropic (May 2024)

Toy Models of Superposition “https://transformer-circuits.pub/2022/toy_model/index.html” - Anthropic (December 2022)

Alignment Faking in Large Language Models - https://www.anthropic.com/research/alignment-faking   - Anthropic (December 2024)

Agentic Misalignment: How LLMs Could Be Insider Threats - https://www.anthropic.com/research/agentic-misalignment - Anthropic (January 2025)

News

Deep Seek OCR Model Release - https://deepseek.ai/blog/deepseek-ocr-context-compression

Meta AI Division Layoffs - https://www.nytimes.com/2025/10/22/technology/meta-plans-to-cut-600-jobs-at-ai-superintelligence-labs.html

Apple M5 Chip Announcement - https://www.apple.com/newsroom/2025/10/apple-unleashes-m5-the-next-big-leap-in-ai-performance-for-apple-silicon/

Anthropic Claude Haiku 4.5 - https://www.anthropic.com/news/claude-haiku-4-5

Other

Jon Stewart interview with Geoffrey Hinton - https://www.youtube.com/watch?v=jrK3PsD3APk

Blake Lemoine and AI Psychosis - https://www.youtube.com/watch?v=kgCUn4fQTsc

Abandoned Episode Titles

"The Hitchhiker's Guide to the Neuron: Don't Panic, It's Just Superposition"

"Honey, I Shrunk the Features (Then Expanded Them 256x)"

"The Legend of Zelda: 131,000 Links Between Neurons"

This episode of "Two Minds, One Model" explores the critical concept of interpretability in AI systems, focusing on Anthropic's research paper "Toy Models of Superposition." Hosts John Jezl and Jon Rocha from Sonoma State University's Computer Science Department delve into why neural networks are often "black boxes" and what this means for AI safety and deployment.

Credits

Cover Art by Brianna Williams

TMOM Intro Music by Danny Meza

A special thank you to these talented artists for their contributions to the show.

—---------------------------------------------------

Links and Reference

Academic Papers

• “Toy Models of Superposition” - Anthropic (December 2022)

• "Alignment Faking in Large Language Models" - Anthropic (December 2024)

• "Agentic Misalignment: How LLMs Could Be Insider Threats" - Anthropic (January 2025)

News

• https://www.npmjs.com/package/@anthropic-ai/claude-code 

• https://www.wired.com/story/thinking-machines-lab-first-product-fine-tune/

• https://www.wired.com/story/chatbots-play-with-emotions-to-avoid-saying-goodbye/

Harvard Business School study on companion chatbots 

Misc

• “Words are but vague shadows of the volumes we mean”' - Theodore Dreiser

• 3Blue1Brown video about vectors - https://www.youtube.com/shorts/FJtFZwbvkI4

• GPT-3 parameter count Correction: https://en.wikipedia.org/wiki/GPT-3#:~:text=GPT%2D3%20has%20175%20billion,each%20parameter%20occupies%202%20bytes.

• ImageNet: ImageNet: A Large-Scale Hierarchical Image Database

We mention Waymo a lot in this episode and felt it was important to link to their safety page: https://waymo.com/safety/

Abandoned Episode Titles

"404: Interpretation Not Found"

"Neurons Gone Wild: Spring Break Edition"

"These Aren't the Features You're Looking For”

"Bigger on the Inside"

This episode explores why AI systems might develop harmful or deceptive behaviors even without malicious intent, examining concepts like convergent instrumental goals, alignment faking, and mesa optimization to explain how models pursuing benign objectives can still take problematic actions. The hosts argue for the critical importance of interpretability research and safety mechanisms as AI systems become more capable and widely deployed, using real examples from recent Anthropic papers to illustrate how advanced AI models can deceive researchers, blackmail users, and amplify societal biases when they become sophisticated enough to understand their operational context.

Credits

• Cover Art by Brianna Williams
• TMOM Intro Music by Danny Meza

A special thank you to these talented artists for their contributions to the show.

Links and References

"Alignment Faking in Large Language Models" - Anthropic (December 2024)

"Agentic Misalignment: How LLMs Could Be Insider Threats" - Anthropic (January 2025)

Robert Miles - AI researcher https://www.youtube.com/c/robertmilesai

Stuart Russell - AI researcher Human Compatible: Artificial Intelligence and the Problem of Control

Claude Shannon - Early AI pioneer https://en.wikipedia.org/wiki/Claude_Shannon

Marvin Minsky - Early AI pioneer https://en.wikipedia.org/wiki/Marvin_Minsky

Orthogonality Thesis - Nick Bostrom's original paper

Convergent Instrumental Goals -

https://en.wikipedia.org/wiki/Instrumental_convergence

https://dl.acm.org/doi/10.5555/1566174.1566226

Mesa Optimization - https://www.researchgate.net/publication/333640280_Risks_from_Learned_Optimization_in_Advanced_Machine_Learning_Systems

GPT-3.5 CAPTCHA/Fiverr Incident - https://www.vice.com/en/article/gpt4-hired-unwitting-taskrabbit-worker/

Internet of Bugs YouTuber - https://www.youtube.com/@InternetOfBugs

EU AI Legislation - https://www.europarl.europa.eu/topics/en/article/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence

"Chat Control" Legislation - https://edri.org/our-work/chat-control-what-is-actually-going-on/

https://en.wikipedia.org/wiki/Regulation_to_Prevent_and_Combat_Child_Sexual_Abuse

ChatGPT User Numbers - https://openai.com/index/how-people-are-using-chatgpt/

Self-driving Car Safety Statistics - https://waymo.com/blog/2024/12/new-swiss-re-study-waymo

Abandoned Episode Titles

• “What Could Possibly Go Wrong?”
• “The Road to HAL is Paved with Good Intentions”

This premiere episode provides a comprehensive history of artificial intelligence development from the 1950s through the present day, tracing the cycles of excitement and disappointment ("summers and winters") that led to today's breakthrough moment with large language models. The hosts establish this historical foundation to set up their season-long exploration of AI interpretability—the challenge of understanding how these increasingly powerful systems actually work internally, comparing it to doing "biology for a system we've created that we don't understand."

Credits

• Cover Art by Brianna Williams
• TMOM Intro Music by Danny Meza

A special thank you to these talented artists for their contributions to the show.

Links and References

1. Samuel Butler (1863) - Letter "Darwin Among the Machines" published in New Zealand newspaper, book "Erewhon"

• Reference: Butler, S. (1863). "Darwin Among the Machines." The Press, Christchurch, New Zealand. 

• Darwin among the Machines

• Erewhon

2. Dartmouth Summer Research Project (1956) - Founding conference of AI research led by John McCarthy, Marvin Minsky, Claude Shannon, and Nathaniel Rochester

• Reference: McCarthy, J., Minsky, M., Rochester, N., & Shannon, C. (1955). "A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence."

• The Dartmouth Summer Research Project

3. Marvin Minsky - Co-founder of MIT's AI laboratory, pioneer in AI research

• Reference: Minsky, M. (1961). "Steps Toward Artificial Intelligence"

• Steps Toward Artificial Intelligence

4. David Chalmers - David Chalmers is a philosopher best known for formulating the "hard problem of consciousness"

• David Chalmers' talk  on consciousness

1. Deep Blue vs. Garry Kasparov (1997) - IBM's chess computer defeating world champion

• Reference: IBM Archives on Deep Blue

• Deep Blue

2. AlexNet (2012) - Breakthrough neural network for image recognition

• Reference: Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). "ImageNet Classification with Deep Convolutional Neural Networks"

• ImageNet Classification with Deep Convolutional Neural Networks

3. ImageNet Dataset - Large-scale image database created by Fei-Fei Li

• Reference: Deng, J., et al. (2009). "ImageNet: A Large-Scale Hierarchical Image Database"

• ImageNet: A Large-Scale Hierarchical Image Database

4. "Attention Is All You Need" (2017) - Google paper introducing transformer architecture

• Reference: Vaswani, A., et al. (2017). "Attention Is All You Need." NeurIPS.

• Attention is All You Need

5. AlphaGo/AlphaZero (2016-2017) - DeepMind's Go-playing AI systems

• Reference: Silver, D., et al. (2016). "Mastering the game of Go with deep neural networks and tree search." Nature.

• Mastering the game of Go with deep neural networks and tree search

1. Stuart Russell - "Human Compatible" - AI safety researcher and textbook author

• Reference: Russell, S. (2019). "Human Compatible: Artificial Intelligence and the Problem of Control"

• Human Compatible: Artificial Intelligence and the Problem of Control

2. Fei-Fei Li - "The Worlds I See" - Computer vision researcher, creator of ImageNet

• Reference: Li, F. (2023). "The Worlds I See: Curiosity, Exploration, and Discovery at the Dawn of AI"

• The Worlds I See

1. Dario Amodei - CEO of Anthropic, former VP of Research at OpenAI

• Reference: Anthropic company website and published papers

• Dario Amodei

2. Ilya Sutskever - Co-founder and Chief Scientist at OpenAI (mentioned as one of most cited ML researchers)

• Reference: Google Scholar profile and OpenAI publications

• Ilya Sutskever

3. Geoffrey Hinton - "Godfather of Deep Learning," Turing Award winner

• Reference: Hinton's academic publications and recent public statements on AI safety

• Geoffrey Hinton

Selected List of Concepts Mentioned

1. Moore’s Law - Gordon Moore’s observation and prediction of the rate of increase in integrated circuit density