Applying Open Source Methods to Building and Training Large Language Models
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only! Grab it Explore an innovative project that applies open source methodologies to large language model development in this 30-minute conference talk by Carol Chen from Red Hat and Mark Sturdevant from IBM. Learn about a groundbreaking approach that enables contributors to enhance existing models without the need for complete forking and fine-tuning. Discover how this technology aims to establish an upstream community built on contributing acceptance workflows, making open source AI more accessible. Gain insights into the process of incrementally adding domain knowledge to models in a unified and open manner, while reducing variations and improving overall performance. Understand the potential impact of this approach on the future of collaborative AI development and the democratization of large language models.