| Description: |
Chatbots like ChatGPT may seem magical, but what’s really going on inside their “brains”? The answer is Transformers—a breakthrough in artificial intelligence. Unlike older models that read words one at a time, transformers look at an entire sentence (or even a whole paragraph) at once, figuring out how each word relates to the others. This “attention” trick allows large language models (LLMs) to generate smooth, meaningful text, translate languages, and even hold conversations that feel natural. In this talk, we’ll break down the magic behind transformers using simple examples, explore how models learn from huge amounts of data, and play with the “settings” that shape their personality—like temperature (how creative the AI sounds) and top-k (how many word options it considers). By the end, you’ll understand not only how these models work, but also why they’re changing the way we study, work, and create. |