Over The Horizon
Get the latest international news and world events from around the world.
Time Crystal in motion
Physicists at CU Boulder have, for the first time, used liquid crystals to create a new kind of time crystal, a curious phase of matter in which particles are in constant motion.
Upgrading an RTX 4090 to 48 GB of VRAM is (slightly) easier than I thought, thanks to a custom PCB, some second-hand GDDR6X, and a leaked Nvidia BIOS
Yet another episode of: “Andy learns things with you.”
A Systems View of LLMs on TPUs
Training LLMs often feels like alchemy, but understanding and optimizing the performance of your models doesn’t have to. This book aims to demystify the science of scaling language models: how TPUs (and GPUs) work and how they communicate with each other, how LLMs run on real hardware, and how to parallelize your models during training and inference so they run efficiently at massive scale. If you’ve ever wondered “how expensive should this LLM be to train” or “how much memory do I need to serve this model myself” or “what’s an AllGather”, we hope this will be useful to you.