lm_eval/tasks/mmlu/continuation/mmlu_security_studies.yaml · 9bd948df0d4d2f3503bbcbf2e78091cf2881e1fa · gaoqiong / lm-evaluation-harness

"tests/models/marian/test_modeling_flax_marian.py" did not exist on "71356034232cfce20d8f24df1213944665e7fe08"

add task for mmlu evaluation in arc multiple choice format (#1745) · 9097ad3e

jonabur authored May 08, 2024



* add mmlu arc style evaluation

* rename arc_style to continuation

---------
Co-authored-by: Jonathan Burdge <jburdge@mahti-login11.mahti.csc.fi>
Co-authored-by: Jonathan Burdge <jburdge@mahti-login12.mahti.csc.fi>

9097ad3e

mmlu_security_studies.yaml 259 Bytes

Replace mmlu_security_studies.yaml