From the course: The AI-Driven Software Developer: Optimize, Innovate, Transform

Unlock this course with a free trial

Join today to access over 24,800 courses taught by industry experts.

Understanding AI models: LLMs, SLMs, open source, and more

Understanding AI models: LLMs, SLMs, open source, and more

From the course: The AI-Driven Software Developer: Optimize, Innovate, Transform

Start my 1-month free trial Buy for my team

Understanding AI models: LLMs, SLMs, open source, and more

“

- [Speaker] A large component of getting the most out of generative AI is understanding large language models or language models in general. If you look over the definition of a language model, you'll likely come across something that sounds like this, probability distribution over language input. Now, this is true, but what does that really mean? At their simplest form, language models will receive a text input or prompt and will continuously try to guess the next likely token or small unit of meaning. That could be a word or perhaps a part of a word. Now, let's look at this example. Define fib. So it's starting of a python function, and then the language model will try to come up with what the probabilities are for different things to be next. So we could have perhaps an 80% chance of n, and then we could have 15% chance of num, and then a 5% chance of number, as what this function receives as its parameter, which would be the next token. The language model will create these…

Contents