GitHub – Karpathy/LLM101n: LLM101n: Let's Build a Storyteller (github.com) 57 points by bilsbie 2y ago ↗ HN
[–] 0x303 2y ago ↗ This appears to just be a README and nothing else [–] jwitthuhn 2y ago ↗ For anyone who wants a sneak peak at what the content will be, Andrej Karpathy already has a series on videos on youtube that covers roughly the first half of the list here: https://www.youtube.com/watch?v=VMj-3S1tku0&list=PLAqhIrjkxb...Starts with building micrograd to build an understanding of how pytorch understands how to calculate gradients, then it proceeds all the way to making a gpt-2 clone.Looks like this is an effort to reorganize and build on that existing work.
[–] jwitthuhn 2y ago ↗ For anyone who wants a sneak peak at what the content will be, Andrej Karpathy already has a series on videos on youtube that covers roughly the first half of the list here: https://www.youtube.com/watch?v=VMj-3S1tku0&list=PLAqhIrjkxb...Starts with building micrograd to build an understanding of how pytorch understands how to calculate gradients, then it proceeds all the way to making a gpt-2 clone.Looks like this is an effort to reorganize and build on that existing work.
[–] stavros 2y ago ↗ I made something similar, which works really well:https://deepdreams.stavros.io/Mine includes the source.
[–] bilsbie 2y ago ↗ Get in touch if anyone is interested in teaming up and going through this when it’s realized.I can’t keep motivated by myself. A group would be really fun. We could discuss each video and keep a timeline. [–] jaybhole 2y ago ↗ I’m down. Here’s my discord: rambdr
7 comments
[ 3.4 ms ] story [ 42.4 ms ] threadStarts with building micrograd to build an understanding of how pytorch understands how to calculate gradients, then it proceeds all the way to making a gpt-2 clone.
Looks like this is an effort to reorganize and build on that existing work.
https://deepdreams.stavros.io/
Mine includes the source.
I can’t keep motivated by myself. A group would be really fun. We could discuss each video and keep a timeline.