Archive
Evaluation & Benchmarks

The Eleuther Harness Name

Archive
Easy
100pts0 solves
EleutherAI's open-source framework for evaluating autoregressive language models across hundreds of standard tasks — the de facto academic benchmark harness — is called what?
Show hint
Three words, first is the model class, last is the metaphor of horse-drawn carts.

Archive — no submissions accepted

This challenge is preserved for reference. Play live challenges at /challenges.