4 ms·Simple and Scalable Strategies to Continually Pre-Train Large Language Models1 points by Anon84 3y ago