Latest / Logosofia no Dia a Dia
What "Distribution" Really Means in AI Models
When we say AI models sample from a distribution, what does that actually mean? This episode breaks down the mechanics — token probabilities, high-dimensional vector spaces, and how training data shapes the "shape" of a model's confidence. We explore the tradeoff between narrow specialization and broad competence, how Mixture of Experts really works, and why the same dynamics apply to text and TTS models. If you've ever wondered why some models feel like experts while others feel like generalists, this one's for you. Episode 4669 — myweirdprompts.com/n/4669 (permanent short link…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.