Merge branch 'big-refactor' of https://github.com/EleutherAI/lm-evaluation-harness into toxicity

16c4afc6 · lintangsutawika · 7b376ae1 · 176d5a26 · 7b376ae1 · 7b376ae1
Commit 16c4afc6 authored Aug 03, 2023 by lintangsutawika
8 changed files
--- a/results/xglm/xglm-7.5B/xglm-7.5B_mathematical_reasoning_few_shot_5-shot.json
+++ b/results/xglm/xglm-7.5B/xglm-7.5B_mathematical_reasoning_few_shot_5-shot.json
--- a/results/xglm/xglm-7.5B/xglm-7.5B_pawsx_0-shot.json
+++ b/results/xglm/xglm-7.5B/xglm-7.5B_pawsx_0-shot.json
--- a/results/xglm/xglm-7.5B/xglm-7.5B_xcopa_0-shot.json
+++ b/results/xglm/xglm-7.5B/xglm-7.5B_xcopa_0-shot.json
--- a/results/xglm/xglm-7.5B/xglm-7.5B_xnli_0-shot.json
+++ b/results/xglm/xglm-7.5B/xglm-7.5B_xnli_0-shot.json
--- a/results/xglm/xglm-7.5B/xglm-7.5B_xstory_cloze_0-shot.json
+++ b/results/xglm/xglm-7.5B/xglm-7.5B_xstory_cloze_0-shot.json
--- a/results/xglm/xglm-7.5B/xglm-7.5B_xwinograd_0-shot.json
+++ b/results/xglm/xglm-7.5B/xglm-7.5B_xwinograd_0-shot.json
--- a/scripts/build_benchmark.py
+++ b/scripts/build_benchmark.py
--- a/tests/extra/test_utils.py
+++ b/tests/extra/test_utils.py