I'm building an LLM from scratch
Just posted Episode 2: Training an LLM from Scratch in Pure NumPy!
The goal of the series is to raise as much money as possible for cancer research.
So far, we've raised $18k for charity : )
Check it out and let me know what you think!
A year ago, I made LLMs speedrun Wikipedia. The goal is to go from one Wikipedia page to another using only internal links.
Last year, LLMs took, on average, about 5-10s. Crazy to see Jev take less than 1s.
Obviously, there's more to the story than this. In fact, reading about how the AI field evolved is pretty interesting.
Btw, this is part of a video I just released on building a GPT from scratch in NumPy.
If you're interested, check it out :)
youtube.com/watch?v=s9w3gtgv…
You can check it out here: youtube.com/watch?v=s9w3gtgv…
Also, if any other companies are interested in sponsoring, DM me
I'm building a GPT from scratch
Just posted Episode 1: Building a GPT with only NumPy
Let me know what you think :)
Thanks @real_deep_ml for sponsoring the episode!!



