Day 2/30 of building a reasoning model from scratch 🤖 today I set up free GPU compute on Lightning AI Studio, walked through round-trip tokenization with the Qwen3 tokenizer to show how text becomes token IDs and back, and loaded the Qwen3-0.6B model to look at what’s actually inside: embeddings, transformer blocks, and everything in between. Following Sebastian Raschka’s Build a Reasoning Model From Scratch - Comment link for link to the repository. #LLM #buildinpublic #reasoningmodels