You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Huggingface: It appears that huggingface is doing evaluations of LLMs in their backend in the form of leaderboard. Is it possible to democratize that? Are there ways for huggingface or the open source community to make the LLMs evaluation process more transparent and make the results more easily reproducible?
The text was updated successfully, but these errors were encountered:
Geekbench: https://browser.geekbench.com/v6/cpu/singlecore
Huggingface: It appears that huggingface is doing evaluations of LLMs in their backend in the form of leaderboard. Is it possible to democratize that? Are there ways for huggingface or the open source community to make the LLMs evaluation process more transparent and make the results more easily reproducible?
The text was updated successfully, but these errors were encountered: