@DanKornas: Building an LLM from scratch is easier when each layer has its own notebook. EveryonesLLM is a Google Colab-based tutor…

Building an LLM from scratch is easier when each layer has its own notebook. EveryonesLLM is a Google Colab-based tutorial repo for building a nanoGPT-style LLM from scratch. It helps you learn the stack step by step by splitting the path into numbered chapters, from dataloading and embeddings to attention, training, scaling, instruction tuning, and a beta vision LLM track. Key features: • Colab-first lessons – each chapter links to an “Open in Colab” notebook with an estimated time • Layer-by-layer path – covers dataloaders, token/position embeddings, LayerNorm, attention, transformer blocks, logits, and nanoGPT • Training progression – includes CPU/GPU speed checks, larger models/datasets, LR schedule, checkpoints, pretraining, and instruction tuning • Exercise + answer notebooks – separate todo and answer notebooks make it easier to practice, then compare • Extra learning aids – README links a tensor map, demo/web app, and beta vision pretraining/instruction-tuning chapters It’s open-source (MIT license). Link in the reply