HN Reader

Link seems to be broken: content briefly loads then is replaced with "Something Went Wrong" then "D is not a function". Stays broken with adblock disabled.

Thanks for sharing; you clearly spent a lot of time making this easy to digest. I especially like the tokens-to-embedding visualisation.

I recently had some trouble converting a HF transformer I trained with PyTorch to Core ML. I just couldn’t get the KV cache to work, which made it unusably slow after 50 tokens…

This is a surprising good read of how LLM works in general.