-
Notifications
You must be signed in to change notification settings - Fork 547
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Examples] Finetune Falcon 7B and 40B Example #2242
[Examples] Finetune Falcon 7B and 40B Example #2242
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Awesome work @xzrderek! Left some comments. Will follow up offline.
Note: one thing to look at in the future https://github.com/AdrianBZG/LLM-distributed-finetune |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @xzrderek! Doing a run now. Left some comments.
…ypilot into derek-falcon-example # Conflicts: # llm/falcon/README.md
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is awesome! We can have a serving yaml for falcon as well, especially the latest falcon-180B, so people can play with the largest model open-source LLM themselves. ; ) |
I've added an example on how to fine-tune Falcon-7B-sharded (for those with less powerful GPUs), Falcon-7B, and Falcon-40B using SkyPilot and wrote a
README.md
(heavily borrowed from the Vicuna README).Ideally, the repo referenced in my code,
https://github.com/xzrderek/sky-falcon.git
, should be a repo under skypilot-org, similar to the sky-llama example, where their code is stored here.