DataHour: LLM Training Tips & Tricks

Explore the latest innovations in natural language processing (NLP) propelled by Large Language Models (LLMs) such as the GPT series. With hundreds of billions of parameters, these models achieve near or comparable human-level performance across various language tasks, often without the need for task-specific training. Dive into an enlightening session with Shaan, where he delves into effective strategies and essential practices for fine-tuning LLMs to unleash their full potential.

Shaan Khosla is a Senior Data Scientist at Nebula where he researches, designs, and develops NLP models. In addition to his primary role, Shaan contributes as a writer for "Let's Talk Text," a newsletter summarizing the latest NLP models, algorithms, and techniques into technical bite-sized chunks. His expertise in the field is further highlighted by his previous tenure at Bank of America, where he was part of an internal machine learning consulting team that used LLMs to build proof of concept systems. Shaan holds a master's degree in Data Science from NYU and a BSBA in Computer Science and Finance from the University of Miami. He has published multiple peer-reviewed papers applying LLMs, topic modeling, and recommendation systems to the fields of biochemistry and healthcare.