(Day 62) Stanford CS224N (NLP with DL) - Transformers, Pretraining, RLHF

Ivan Ivanov · March 3, 2024

Hello :) Today is Day 62!

A quick summary of today:

I watched this State of GPT talk by Andrej Karpathy a week or so ago and saw the below slide:

image

And that was the first time I saw this GPT pipeline. Today, I got to learn about each of these steps. Feels nice knowing how GPT/ChatGPT works.

Anyway ~ my notes are below :)

Lecture 8: Self-Attention and Transformers

image image image image image image image image image image image image

Lecture 9: Pretraining

image image image image image image image image image image image

Lecture 10: Prompting, Reinforcement Learning from Human Feedback

image image image image image image image image

Tomorrow we continue diving deeper into the world of NLP.

That is all for today!

See you tomorrow :)