ASAdi Shikinopeningthehood.hashnode.dev·3d ago · 11 min readThe Inside of a Model - Part B: How a Model LearnsIn this post, we finally get to the question that makes neural networks feel almost magical: how does a model actually “learn”? Small recap from the last post: we talked about MLPs and how they proces00
ASAdi Shikinopeningthehood.hashnode.dev·Jul 20 · 9 min readInside the Model, Part I: From Vector to PredictionWe all heard about neural networks. The idea sounds exciting: make a machine that thinks like a human. I hate to be the one to break it down, but neural networks do not really “think” like humans. At 00
ASAdi Shikinopeningthehood.hashnode.dev·Jul 13 · 11 min readWhen the Embedding Is Wrong, Everything After It Is WrongIn the last post, I wrote about fine-tuning and why I think the next wave of applied AI will be about moving from using models to shaping them. This time, I want to go one level deeper into a real use00
ASAdi Shikinopeningthehood.hashnode.dev·Jul 7 · 8 min readAPIs Were the First Layer. Training Is the Next OneWe all see the world changing. A few years ago, AI was not really accessible to most builders. The important research existed long before ChatGPT, including the 2017 paper Attention Is All You Need, w00
