All AI Models Might Be The Same

what can language model embeddings tell us about whales speech, and decoding ancient texts? (on The Platonic Representation Hypothesis and the idea of universality in AI models)

Read in full here: