stanford-crfm/helm

Official Llama 3.1 Evals

YLGH opened this issue · 2 comments

YLGH commented

Hey,

Is anyone currently working on adding the official Meta llama 3.1 eval dataset/tasks to helm?
https://huggingface.co/datasets/meta-llama/Meta-Llama-3.1-405B-Instruct-evals

Thanks!

Thanks for the interest! We are working on the Llama 3.1 Instruct evaluations currently and will have them soon.

Llama 3.1 Instruct Turbo evaluations have been released! (Lite, MMLU)