Reading up on PyTorch
2 deep · digging since may 09
- LLMs are complicated now – Ian’s Blog
Modern LLMs have grown complex with many attention variants and mixture-of-experts, echoing the messy evolution of recommendation systems.
- Train Your Own LLM from Scratch
A hands-on workshop guides readers through building a ~10M-parameter GPT model from scratch in PyTorch, trainable on a laptop in under an hour.