Past Event: Using LLMs on Yale's computing clusters- hands on workshop

This workshop will provide researchers interested in LLMs with the skills to launch and run inference-based workflows with open-source models on YCRC HPC systems using Ollama.

Specifically, attendees will learn: 

• Advantages of using YCRC systems for LLMs
• How to identify what GPU is needed for different LLMs 
• How to launch an LLM on YCRC systems via the terminal and perform basic inference
• How to modify an LLM's parameters for reproducibility/consistency/creativity in response 
• How to implement an LLM within python using jupyter and ollama 
• Initial setup for huggingface environments
• Additional considerations for RAG and fine-tuning

Prerequisites: Introductory-level knowledge of Yale's HPC clusters. Recommended course: https://research.computing.yale.edu/intro-to-hpc