llama-2-70b model (open source from Meta) = 140GB params + ~500 lines of .c code (!!!)
[1hr Talk] Intro to Large Language Models - YouTube
Slides as PDF: https://drive.google.com/file/d/1pxx_... (42MB)
The Unreasonable Effectiveness of Recurrent Neural Networks (from 2015)
PhD from Stanford in AI/ML (CNNs)
Lead of self-driving in Tesla
Co-founder of OpenAI
facebookresearch/llama: Inference code for LLaMA models @GitHub
book: Thinking, Fast and Slow by Kahneman, Daniel
system 1 vs system 2 "thinking"
LLMs are currently only "system 1" (fast and unreliable)
future: attempt "system 2 thinking", "trade time for accuracy", take more time for better result.
"self improvement": hard in language domain, no "rules"
"AI apps store"
No comments:
Post a Comment