News for hackers. Stuff that matters.

A comment on Show HN: Llama 3.2 Interpretability with Sparse Autoencoders. Parent comment

Re:Show HN: Llama 3.2 Interpretability with Sparse Autoencoders by stavros on Friday November 22, 2024 @05:18AM (#42212586)
Then I think you'll be very disappointed. LLMs aren't in the same category as calculators, for example.
[ Parent | on HN ]
Re:Show HN: Llama 3.2 Interpretability with Sparse Autoencoders by f_devd on Friday November 22, 2024 @07:06AM (#42213145)
I have no illusions on LLMs, I have been working with them since og BERT, always with these same issues and more. I'm just stating what would be needed in the future to make them reliably useful outside of creative writing & (human-guided & checked) search.

If an LLM provides an incorrect/orthogonal rhetoric without a way to reliably fix/debug it it's just not as useful as it theoretically could be given the data contained in the parameters.

[ Parent | on HN ]