BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CERN//INDICO//EN BEGIN:VEVENT SUMMARY:Foundations of LLM Mastery: Fine-tuning on multi GPUs DTSTART:20250911T070000Z DTEND:20250911T103000Z DTSTAMP:20260718T180600Z UID:indico-event-204@events.asc.ac.at CONTACT:training@asc.ac.at DESCRIPTION:Start date:  11 Sept 2025\, 09:00 (CEST)     Entry level:  IntermediateEnd date:  11 Sept 2025\, 12:30 (CEST)   Subject area:   Artificial IntelligenceLocation:  ONLINE (Zoom)     Topics:  Large La nguage ModelsLanguage:  English     Target audience:  Industry\, publ ic sector\, academiaPrice:  Free (for eligible participants)     Organ izers:  ASC & EuroCC Austria & AI:ATScaling fine-tuning of large language models (LLMs) to multiple GPUs can unlock new levels of performance and e fficiency\, making it accessible for industries of all sizes.In this 3.5-h our course\, participants from start-ups\, SMEs\, and large enterprises wi ll gain hands-on experience with powerful multi-GPU fine-tuning techniques \, optimizing their LLM workflows for both speed and scalability.Key topic s include:Distributed Data Parallel (DDP): Efficiently distribute data and model updates across multiple GPUs.ZeRO with DeepSpeed: Scale large model s while managing memory effectively.Fully Sharded Data Parallel (FSDP): Ye t another technique to reduce memory usage by sharding.Hugging Face Accele rate: Simplify multi-GPU training with intuitive tools and configurations. Through guided exercises and real-world examples\, this course provides pa rticipants with the skills to fine-tune LLMs efficiently across multiple G PUs\, preparing them to tackle complex\, large-scale projects.By the end o f this course\, participants will have developed a foundational understand ing of techniques to distribute model parameters and states accross GPUs o n several nodes.Agenda & ContentFor a detailed timetable and additional in formation\, please see Agenda & Content in the left menu.Entry levelInterm ediatePrerequisitesBasic programming skills in Python.                       Having worked with Pytorch would be beneficial.Tar get audienceCourse for academia\, industry\, and public administration.Cou rse formatThis course will be delivered as a LIVE ONLINE COURSE (using Zoo m).Hands-on labsAll participants will get a temporary user account on LEON ARDO to do the hands-on labs.You will use your own laptop or workstation t o connect conveniently from your browser to a LEONARDO Jupyterlab session and do the hands-on exercises.LecturersSimeon Harrison (INiTS and EuroCC A ustria)Martin Pfister (ACA – Advanced Computing Austria GmbH and EuroCC Austria)LanguageEnglishDate\, time\, and location11.09.2025\, 09:00 – 12 :30 CEST\, LIVE ONLINE COURSE (using Zoom)Prices and eligibilityThis cours e is partly funded by the EuroCC 2 project. Therefore\, the course is open and free of charge for participants from academia\, industry\, and public administration from the Member States (MS) of the European Union (EU) and Associated/Other Countries to the Horizon 2020 programme.RegistrationThe registration form and deadlines are available from the left menu. The regi stration form might close even before the deadline if the number of regist rations has reached its maximum (and will open again if there are cancelat ions).Please register with your official email address to prove your affil iation.You will get an automatic confirmation via email (subject starting with "[Indico] Registration")\, please check your Spam/Junk folders.Follow ing your successful registration\, you will receive further information ap proximately 1 week before the course.WaitinglistAfter the number of regist rations has reached its maximum or the registration form has been closed\, you may want to send us an email stating that you are interested to be pu t on the waiting list (vacancies may occur due to cancelations\, etc.).To be able to do the hands-on labs on the ASC please provide your full intern ational mobile-phone number for the two-factor authentication required to login to the ASC systems.Modification\, withdrawal & no-show policyYour re gistration is binding. Please only register for the course if you are real ly going to attend.You can update your registration data or withdraw your registration anytime before the registration form has been closed via the link "Manage my registration" which you can find at the bottom of your aut omatic email confirmation (subject starting with "[Indico] Registration"). Alternatively\, or after the registration form has been closed\, please in form us about your cancelation or any change in your registration data (es pecially your mobile-phone number) via email (training@asc.ac.at).No-show policy: If you do not cancel and do not show up at the course you will be blacklisted and excluded from future training events.Organizers                This course is jointly organized by the ASC Research Cent er\, TU Wien\, EuroCC Austria\, and AI:AT.AcknowledgementsThis course is p artially funded by the EuroCC 2 project.The project has received funding f rom the European High-Performance Computing Joint Undertaking (JU) under g rant agreement No 101101903. The JU receives support from the Digital Euro pe Programme and Austria\, Belgium\, Bulgaria\, Croatia\, Cyprus\, Czech R epublic\, Denmark\, Estonia\, Finland\, France\, Germany\, Greece\, Hungar y\, Iceland\, Ireland\, Italy\, Latvia\, Lithuania\, Luxembourg\, Monteneg ro\, Netherlands\, North Macedonia\, Norway\, Poland\, Portugal\, Romania\ , Serbia\, Slovakia\, Slovenia\, Spain\, Sweden\, and Türkiye. Additiona l funding for the project comes from the Austrian federal ministries BMBWF and BMK.\n\nhttps://events.asc.ac.at/event/204/ URL:https://events.asc.ac.at/event/204/ END:VEVENT END:VCALENDAR