A arxiv.org Better and Faster Large Language Models via Multi-Token Prediction 302 points jasondavies 2 years ago 128 comments
