Support upload metadata to hf

allenai / reward-bench

RewardBench: the first evaluation tool for reward models.

Apache License 2.0

440 stars 52 forks source link

Closed vwxyzjn closed 1 month ago

vwxyzjn commented 2 months ago

Quality of life improvements: It directly pushes the eval results to be visible

vwxyzjn commented 1 month ago

Done. Let's just record as much as possible.

Feel free to merge after this.