The moment we stopped understanding AI [AlexNet]

Thanks to KiwiCo for sponsoring today's video! Go to www.kiwico.com/welchlabs and use code WELCHLABS for 50% off your first month of monthly lines and/or for 20% off your first Panda Crate.

Activation Atlas Posters!
www.welchlabs.com/resources/5gtnaauv6nb9lrhoz9cp60…
www.welchlabs.com/resources/activation-atlas-poste…
www.welchlabs.com/resources/large-activation-atlas…
www.welchlabs.com/resources/activation-atlas-poste…

Special thanks to the Patrons:
Juan Benet, Ross Hanson, Yan Babitski, AJ Englehardt, Alvin Khaled, Eduardo Barraza, Hitoshi Yamauchi, Jaewon Jung, Mrgoodlight, Shinichi Hayashi, Sid Sarasvati, Dominic Beaumont, Shannon Prater, Ubiquity Ventures, Matias Forti

Welch Labs
Ad free videos and exclusive perks: www.patreon.com/welchlabs
Watch on TikTok: www.tiktok.com/@welchlabs
Learn More or Contact: www.welchlabs.com/
Instagram: www.instagram.com/welchlabs
X: twitter.com/welchlabs

References
AlexNet Paper
proceedings.neurips.cc/paper_files/paper/2012/file…

Original Activation Atlas Article- explore here - Great interactive Atlas! distill.pub/2019/activation-atlas/
Carter, et al., "Activation Atlas", Distill, 2019.

Feature Visualization Article: distill.pub/2017/feature-visualization/
`Olah, et al., "Feature Visualization", Distill, 2017.`

Great LLM Explainability work: transformer-circuits.pub/2024/scaling-monosemantic…
Templeton, et al., "Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet", Transformer Circuits Thread, 2024.

“Deep Visualization Toolbox" by Jason Yosinski video inspired many visuals:
   • Deep Visualization Toolbox  

Great LLM/GPT Intro paper
arxiv.org/pdf/2304.10557

3B1Bs GPT Videos are excellent, as always:
   • Attention in transformers, step-by-step | DL6  
   • Transformers (how LLMs work) explained vis...  

Andrej Kerpathy's walkthrough is amazing:
   • Let's build GPT: from scratch, in code, sp...  

Goodfellow’s Deep Learning Book
www.deeplearningbook.org/

OpenAI’s 10,000 V100 GPU cluster (1+ exaflop) news.microsoft.com/source/features/innovation/open…

GPT-3 size, etc: Language Models are Few-Shot Learners, Brown et al, 2020.

Unique token count for ChatGPT: cookbook.openai.com/examples/how_to_count_tokens_w…

GPT-4 training size etc, speculative:
patmcguinness.substack.com/p/gpt-4-details-reveale…
www.semianalysis.com/p/gpt-4-architecture-infrastr…

Historical Neural Network Videos
   • Convolutional Network Demo from 1989  
   • Perceptron Research from the 50's & 60's, ...  

Errata
1:40 should be: "word fragment is appended to the end of the original input". Thanks for Chris A for finding this one.