LMFingerprints: Visual Explanations of Language Model Embedding Spaces through Layerwise Contextualization Scores
Sign Up to like & getrecommendations! Published in 2022 at "Computer Graphics Forum"
DOI: 10.1111/cgf.14541
Abstract: Language models, such as BERT, construct multiple, contextualized embeddings for each word occurrence in a corpus. Understanding how the contextualization propagates through the model's layers is crucial for deciding which layers to use for a… read more here.
Keywords: contextualization; visual explanations; embedding spaces; explanations language ... See more keywords