At my current role, I have been building a few Agentic workflows using LLMs. While the work has mostly revolved around practical applications of GenAI, the process has made me curious about learning how the LLM model work internally. So I watched a couple of videos to skim the basic ideas involved. But I want to dive deeper - hopefully implement a small toy project.
Thus, I have picked up a book recommended to me by a friend - Build an LLM from Scratch by Sebastian Raschka1. The author recorded a complimentary video series that covers a lot of the concepts from the book.
Though I’m not sure about the cadence to progress through the book, I would be happy if I can learn a bit and wrap it up by the end of the year.