​Details

​A team member from Aleph Alpha Research will talk about the fundamentals of LLM training.

​Abstract: Scale is one key ingredient that has enabled the success of modern LLMs. In this talk I will go over different kinds of parallelisms (model, pipeline, data, sequence, expert) and how they compose with each other to enable training of models with extremely large number of parameters.

​🔈 Speaker: Vedant Nanda

​Agenda:
✨ 18:30 Doors open: time for networking with fellow attendees
✨ 19:00 Talk and Q&A
✨ 20:00 Mingling and networking with pizza and drinks
✨ 21:00 Meetup ends

​Where: In person, Aleph Alpha Heidelberg, Speyerer Str. 14, 69115 Heidelberg
When: July 22, 2025
Language: English

​⚠️ Registration is free, but required due to building security.