Volodymyr Barannik comments on Mechanistic Interpretability of Llama 3.2 with Sparse Autoencoders

[Volodymyr Barannik]

Thanks for your work, but why is the project currently taken down?