LLM Supervised Fine-Tuning Workshop¶

Welcome, and thank you for your interest in this hands-on workshop organized by HES-SO Valais-Wallis and HEIG-VD as part of a Swiss AI Center cheque supported by Exoscale!

This workshop aims at demystifying the process of fine-tuning large language models (LLMs) by guiding you through the entire workflow, using open-source tools and modern cloud infrastructure.

Here is what you will learn:

By the end, you will be able to:

• prepare datasets for training
• train models efficiently from a single GPU to multi-GPU setups
• track training progress in real time
• evaluate performance of fine-tuned models
• deploy your model and even interact with it

Points of contact:

• Andrei Coman (andrei.coman@hevs.ch)
• Pamela Delgado (pamela.delgado@heig-vd.ch)

Contributors:

• Olivier Lemer (olivier.lemer@heig-vd.ch)
• Rémy Marquis (remy.marquis@heig-vd.ch)