EvalEval Coalition

AI & ML interests

We’re building a research coalition on evaluating evaluations (EvalEval)! Hosted by Hugging Face, University of Edinburgh, and EleutherAI.

Recent Activity

EvalEvalBot  updated a dataset about 13 hours ago

evaleval/EEE_datastore

evijit  new activity about 13 hours ago

evaleval/EEE_datastore:[Submission] Fix win_rate scale (0-1) and merge Fibble variants into composite benchmark

EvalEvalBot  new activity about 13 hours ago

evaleval/EEE_datastore:[Submission] Fix win_rate scale (0-1) and merge Fibble variants into composite benchmark

View all activity

Team members 38