juvi

fix: Enable NVIDIA GPU support for Ollama in Docker

2

added NVIDIA GPU support for OLLAMA in docker-compose.yaml

### What piece of documentation is affected? https://github.com/OpenAccess-AI-Collective/axolotl/blob/main/docs/multi-node.qmd ### What part(s) of the article would you like to see updated? Similar to the currently available documentation, it would be great...

In certain areas, I thought the code would benefit from using Einops instead of standard PyTorch operations.