Course Notes — Introduction to Prompt Engineering for Generative AI

Aakash Goel
NLP Experiment
Published in
3 min readMay 30, 2023

--

Link — Joining the NLP Revolution (linkedin.com)

Note: I highly recommend to have first hand experience from Ronnie course only. This article is just having few main content/learnings from it for revision purpose.

What is Generative AI ?

Leverages AI that generate data such as text, images, audio, video, code. Its very fast changing ecosystem.

What is Prompt Engineering ?

Crafting ideal inputs in order to get the most out of LLMs.

Examples of Language Model

Tokens

Small unit that can be easily understood by LLMs. One word can be made up of multiple tokens.

Tokenization

The mechanism by which a model splits its inputs into tokens. can greatly affect model’s output.

What is LLMs ?

Takes an input and produces a token as output.

Zero shot Learning

GPT-3

An ecosystem of LLMs based on 170B parameter model trained on enormous amount of data. You can play with there configuration.

Few-shot Learning

Method for training a model to perform specific tasks.

We are asking the model what it must do before it can insert a stop sequence. The model will attempt to follow the template we’ve provided. Its likely that the model will pick a response from the paragraph.

Text to image

Fine Tuning

Challenge of summarizing for 2nd Grade student

LLMs are very powerful tools, please use them responsibly and create positive impact.

Links to try out

  1. https://platform.openai.com/playground
  2. https://platform.openai.com/examples
  3. https://studio.ai21.com/playground/complete — G1 model by AI21 labs
  4. https://openai.com/product/dall-e-2
  5. https://labs.openai.com/

Please give a clap to this this article if it has helped you. I also welcome your feedback in the Comments section below.

Thanks !!

--

--