I don't have enough VRAM on my local machine to train locally so I followed the documentation on training on Google Colab instead. However, the output of that training was several files (pictured below)

I downloaded the above files by zipping them and downloading the .zip file and then extracting them into a local directory.

From my understanding, the following steps (ilab model test, ilab model serve, ilab model chat) involve working with a GGUF file from a Hugging Face repository, not a .zip file or a collection of files.

Can anyone please provide some direction or instructions on how to convert between the Colab output files and a GGUF file?
Am I mistaken about some part of this process and misleading myself?