How I would learn about LLMs in 2024 (If I had to start over)
Today, Large Language Models have transformed every field. Lawyers, artists, programmers, and many more professionals are utilizing this wonderful technology. In this post, I'll share with you how I would teach myself about LLMs if I had to start all over again.
After a lot of experimentation, integration, and examples, I've decided to write this post to share with you how I'd learn about LLMs if I were to do it all over again.
So, if you're just getting started with LLMs, this post is for you!
Let's break it down:
- Start with the history
- Understand the science
- Fast-forward to the present
- Master prompting techniques
- Learn about embeddings and vector databases
- Play around with data frameworks
- Learn about privacy and security risks
- Become a member of the blog đ
Since I'm a super-fan of instant results, my brain functions as follows:
- I come across an interesting topic.
- I get excited about it.
- I (obviously) skip the docs.
- I start writing code to see how it works.
But since this post is for the wiser me, I'd take a step back and start with the basics...
Let's begin!
Start with the history
I know, history... Boring.
Plus, who cares what happened a thousand years ago... Right?
You know... Nothing fancy, sites like Wikipedia and YouTube offer a wealth of knowledge. The source of information doesn't matter, as long as it's known to be reliable and trustworthy.
History is the best place to start, how else would we get answers to key questions, such as:
- Why were LLMs created?
- What was the motivation?
- What were the challenges and obstacles?
- Who was behind the most important innovations?
- And more...
Answering such questions gives us an idea of how everything came to be and serves as an additional context that we will use at a later time to build a solid foundation about the topic.
