1
Matrix Multiplications on GPUs Run Faster When Given “Predictable” Data (2024)
Great minds discuss flops per watt.
Horace He
ML Systems from first principles. Aims to be better than a ChatGPT summary. Click to read Thonk From First Principles, by Horace He, a Substack publication with thousands of subscribers.
- Popular posts
- 2
- Total points
- 336
- HN comments
- 69
- Rank
- #7,993 personal
Visit thonking.ai → RSS feedOn Hacker News Nov 2024 – May 2026
2
What Shapes Do Matrix Multiplications Like?
Divining order from the chaos