Meta scales Language Model to 1.1T parameters using Mixture of Expertstwitter.com2 points·danielcampos93··0 commentsOpen articleSaveView on HN