A Beginners Guide to Key Large Language Model Terms With Examples

Tanmay Deshpande
3 min readAug 19, 2023
Source — Image by Author

1. Generative AI:

Generative AI models are designed to produce new, original content based on their training data. They can create a variety of outputs, from text and images to music.

Example: Consider a Generative AI trained on classical music. When prompted, it can compose a new piece of music that sounds classical but is entirely original and wasn’t part of its training data.

2. Large Language Model (LLM):

LLMs are trained on vast amounts of text data, enabling them to understand the context and generate human-like text. They can answer questions, write essays, and even generate poetry.

Example: If you ask an LLM, “Write a short poem about the moon,” it might produce:

Silver orb in the night sky,
Casting shadows, shining high.
Guardian of dreams, so bright,
Guiding us with gentle light.

3. Tokens:

In LLMs, a token can represent a character, a word, or anything in between. Tokens are the building blocks the model reads.

Example: The sentence “ChatGPT is fun!” is tokenized into

[“Chat”, “G”, “PT”, “ is”, “ fun”, “!”].

4. Completion Tokens:

--

--

Tanmay Deshpande
Tanmay Deshpande

Written by Tanmay Deshpande

I write about technology in simple words!

No responses yet