Omkaar Kamath
Twitter
LinkedIn
GitHub
Youtube
How to Train an LLM: Part 3
2025-11-13
How to Train an LLM: Part 2
2025-11-13
How to Train an LLM: Part 1
2025-11-11
Cut Cross Entropy > Torch's version? No.
2025-10-16
Cut Cross Entropy from first principles
2025-10-10
Decisive guide on Speculative Decoding
2025-09-15
The Illustrated FastVLM: Apple
2025-09-08
So, why is attention quadratic?
2025-09-05
The Illustrated LFM-2: Liquid AI
2025-09-03
Floating point maths made simple
2025-09-02
Why >> How
2025-08-22
Why I made this website
2025-08-21