|
|
|
|
|
by nelhage
1532 days ago
|
|
(author here) I actually did a v0 writeup in Go, but I wanted a language that had a bit more powerful type system and more support for a fluent/functional style in some of the expressions. I was optimizing for what I know and felt was highly expressive; I've since gotten a bunch of feedback from people who are interested in the content but not comfortable reading Rust, so perhaps it wasn't the best choice. |
|
Aside from that, I thought it was fairly legible. Great write-up by the way. Squashing things into state helps get rid of some of the spookiness created by matrix multiplication and back-propagation. I also really appreciated seeing the explanation on the actual MLP part of the transformer as that is typically assumed to be prior knowledge in other tutorials.