| user: | qwertyforce |
| created: | January 9, 2022 |
| karma: | 773 |
| about: | github.com/qwertyforce |
| 1. | I (Spiritually) Won Comma.ai's Compression Challenge(aaronleslie.dev) |
| 2. | Flash Attention Is Not Always Faster for Short Sequences(blog.qwertyforce.dev) |
| 3. | Fast Attention for Short Sequences(blog.qwertyforce.dev) |
| 4. | Forecast: Stack Overflow Has About 2 Years Left(blog.qwertyforce.dev) |
| 5. | Math and Me(togelius.blogspot.com) |
| 6. | Forecasting the Death of StackOverflow(blog.qwertyforce.dev) |
| 7. | Google AI Studio is now sponsoring Tailwind CSS(twitter.com) |
| 8. | Flash Attention from Scratch(lubits.ch) |
| 9. | Learn Cutlass the Hard Way(kapilsharma.dev) |
| 10. | QK Norm and the Curious Case of Logit Drift(rossjtaylor.com) |
| 11. | Engineering the highest cited cat, Larry(reeserichardson.blog) |
| 12. | ICLR Blogposts 2024(iclr-blogposts.github.io) |
| 13. | |
| 14. | |
| 15. | ICLR Blogposts(iclr-blogposts.github.io) |
| 16. | Why Batch Norm Causes Exploding Gradients [2020](kyleluther.github.io) |
| 17. | Tensor Core Performance: The Ultimate Guide (2020) [pdf](developer.download.nvidia.com) |