From the course: Building a Project with the ChatGPT API

Unlock the full course today

Join today to access over 24,700 courses taught by industry experts.

Examine key concepts

Examine key concepts - ChatGPT Tutorial

From the course: Building a Project with the ChatGPT API

Start my 1-month free trial Buy for my team

Examine key concepts

“

- By now you're familiar with key terms and concepts like models, LLMs, fine tuning, prompts, tokens, completions, embeddings, transformers, and more. Let's pool all these concepts together and introduce a few new concepts like probabilities and completion parameters. At the center, you have foundation LLMs like GPT trained on terabytes of data from the internet, including text, websites, images, online books, and more. Once you have a foundation LLM, you can fine tune it using data for your specific use case. ChatGPT takes a series of prompts that go through a tokenization process by which the model processes the input text by breaking it down into smaller units called tokens. Now I'm going to introduce a new concept, probabilities. It's when the model aims to answer your input prompt by predicting which token is most likely to come next. Take a look at this example using the phrase, which cheese pairs. The model predicts…

Contents