Jupyter notebook issue while learning in short course LLama

I’ve tried restarting the kernel and also clearing my cache. It worked a few days ago but looks like it stopped all of sudden with llama. Also tried Jupyter in other lessons of the course. Same issue. Any advice on how to get Jupyter back up and running?

And I made sure to run the initial
from utils import llama, code_llama

Chrome Browser on Mac

Version 116.0.5845.187 (Official Build) (x86_64)

Trying to see if I can just repl.it side by side with the learning modules but waiting for approval on Hugging Face to install and get permissions to the llama model

I had that exact same error in the first lesson. I ended up creating my own API key on together.ai and updating the url in the llama function in utils.py (the helper function) to url=‘https://api.together.xyz/inference’.

It worked for me after that fix, but the course developers need to do a proper fix I think.

Hope that helps.

1 Like

Several of the Short Courses are broken at the moment since new versions of the LLM’s have been published.

I believe the DLAI staff is working on updates.

1 Like

Thanks for update. I also found in the “How Business Thinkers Can Start Building AI Plugins With Semantic Kernel” , I had to adjust the “Org the tools you make for later use…” lesson, that I had to update the code to a currently supported model,gpt-3.5-turbo in order for code to work. Works great now, but others might have this issue

1 Like