Spaces:

bethgelab
/

lm-similarity

Running

App Files Files Community

lm-similarity / src /utils.py

Commit History

[Add] add bbh and gpqa benchmarks again with correct answer_index selection

0a42e99

Joschka Strueber commited on Feb 7

[Fix] convert logits to softmax for kappa_p

00b5438

Joschka Strueber commited on Feb 5