Stories by jxmorris12

GPT-Fast

Exploring EXIF (2023)

The Practitioner's Guide to the Maximal Update Parameterization

The scientific method and its application to the science of deep learning

Solving Humanity's Last Exam Problems

Why We Think

Philosophical Thoughts on Kolmogorov-Arnold Networks (2024)

Matmul() using PyTorch's MPs back end is faster than Apple's MLX

The Making of Gemini Plays Pokémon

Facebook is not worth $33B (2010)

Comefrom

Diffusion Language Models Are Super Data Learners

How to build a router for MOE models

The Eponymous Principles of Management – Coase's Ceiling and Floor

No One Is Working

No One Is Working

SFT Is Bad RL

A Simple CPU on the Game of Life (2021)

Trends in LLM-Generated Citations on ArXiv

1
2
3
4
5
6
7
8
9
10