MLT __init__ Session #17: LLM int8

MLT __init__ Paper Reading & Discussion 📌 Tim Dettmers, Mike Lewis, Younes Belkada, Luke Zettlemoyer: LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale Presenter: Tim Dettmers Tim is a fifth-year PhD student advised by Luke Zettlemoyer at the University of Washington. He has a background in applied math and automation. His main work focuses on making transformers more efficient and understanding how they work. Personal website: https://timdettmers.com/, Twitter: https://twitter.com/Tim_Dettmers MLT __init__ is a monthly event led by Jayson Cunanan and J. Miguel Valverde where a paper is first presented by a volunteer and then discussed among all attendees. Our goal is to give participants good initializations to study Deep Learning effectively. We also hope to promote collaboration between participants. We will try to achieve this by: ⭐️ Discussing fundamental papers whose key ideas apply to state-of-the-art models ⭐️ Providing the audience with summaries, codes, and visualizations to help understand the critical parts of a research paper Find more information, the videos, and materials from previous sessions in https://github.com/Machine-Learning-Tokyo/init ========================= MLT (Machine Learning Tokyo) site: https://www.mlt.ai/ twitter: https://twitter.com/__MLT__ meetup: https://www.meetup.com/Machine-Learning-Tokyo/ linkedin: https://www.linkedin.com/company/mltai/ github: https://github.com/Machine-Learning-Tokyo slack: https://machinelearningtokyo.slack.com