Description
This webpage contains an interactive resource through the history and mechanics of Language Models — from probabilistic n-grams and word embeddings to transformers, large language models, and the interpretability of what they learn. Not a passive lecture: every concept comes with a manipulable visualization or widget.