The Science of Large Language Models and Language Models for Science
Paul G. Allen School via YouTube
Overview
Explore the cutting-edge developments in large language models and their applications in scientific research through this keynote address from the 2023 Research Showcase. Delve into the Open Language Model (OLMo) project, which aims to create powerful language models that are fully accessible to researchers. Learn about DOLMa, a massive 3T token open dataset for training language models, the instruction-tuned Tulu model, and OLMo v1, a completely open 7B parameter language model. Gain insights into the challenges and importance of open-source AI development, as well as the potential impact of these models on various fields of study. Discover the speaker's expertise in NLP and AI, focusing on understanding, analyzing, and constructing language models, and her contributions to the field through various research projects and accolades.
Syllabus
The science of LLMs and language models for science: Hanna Hajishirzi (Allen School + AI2)
Taught by
Paul G. Allen School