smarty_pants VS neural-network-from-scratch

The idea is that you can simply use this crate with your project to easily train a neural network using your project. The library supports creating, training, parsing, and running. It may gain more functionality in the future. As it stands it's quite small and pretty fast with 5 NeuralNetworks taking nano-seconds to train 1000 generations in the example program. I've tried to make sure that it is "complete" and as such, I've documented nearly every function, method, and struct. I've also written an example project and tried to make it relatively easy to use.

I built a toy neural network that runs in the browser[1] to model 2D functions with the goal of doing something similar to this research (in a much more limited manner, ofc). Since the input space is so much more limited than language models or similar, it's possible to examine the outputs for each neuron for all possible inputs, and in a continuous manner.

In some cases, you can clearly see neurons that specialize to different areas of the function being modeled, like this one: https://i.ameo.link/b0p.png

This OpenAI research seems to be feeding lots of varied input text into the models they're examining and keeping track of the activations of different neurons along the way. Another method I remember seeing used in the past involves using an optimizer to generate inputs that maximally activate particular neurons in vision models[2].

I'm sure that's much more difficult or even impossible for transformers which operate on sequences of tokens/embeddings rather than single static input vectors, but maybe there's a way to generate input embeddings and then use some method to convert them back into tokens.

[1] https://nn.ameo.dev/

[2] https://www.tensorflow.org/tutorials/generative/deepdream

Full source code is on Github: https://github.com/ameobea/neural-network-from-scratch