Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
gaoqiong
lm-evaluation-harness
Commits
bdddfec2
Unverified
Commit
bdddfec2
authored
Sep 21, 2023
by
Hailey Schoelkopf
Committed by
GitHub
Sep 21, 2023
Browse files
Merge pull request #864 from EleutherAI/add-fewshot-config
[Refactor] CMMLU, C-Eval port ; Add fewshot config
parents
0f6cd358
f88ffeee
Changes
131
Hide whitespace changes
Inline
Side-by-side
Showing
20 changed files
with
80 additions
and
0 deletions
+80
-0
lm_eval/tasks/cmmlu/cmmlu_default_high_school_geography.yaml
lm_eval/tasks/cmmlu/cmmlu_default_high_school_geography.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_high_school_mathematics.yaml
...al/tasks/cmmlu/cmmlu_default_high_school_mathematics.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_high_school_physics.yaml
lm_eval/tasks/cmmlu/cmmlu_default_high_school_physics.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_high_school_politics.yaml
lm_eval/tasks/cmmlu/cmmlu_default_high_school_politics.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_human_sexuality.yaml
lm_eval/tasks/cmmlu/cmmlu_default_human_sexuality.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_international_law.yaml
lm_eval/tasks/cmmlu/cmmlu_default_international_law.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_journalism.yaml
lm_eval/tasks/cmmlu/cmmlu_default_journalism.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_jurisprudence.yaml
lm_eval/tasks/cmmlu/cmmlu_default_jurisprudence.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_legal_and_moral_basis.yaml
lm_eval/tasks/cmmlu/cmmlu_default_legal_and_moral_basis.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_logical.yaml
lm_eval/tasks/cmmlu/cmmlu_default_logical.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_machine_learning.yaml
lm_eval/tasks/cmmlu/cmmlu_default_machine_learning.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_management.yaml
lm_eval/tasks/cmmlu/cmmlu_default_management.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_marketing.yaml
lm_eval/tasks/cmmlu/cmmlu_default_marketing.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_marxist_theory.yaml
lm_eval/tasks/cmmlu/cmmlu_default_marxist_theory.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_modern_chinese.yaml
lm_eval/tasks/cmmlu/cmmlu_default_modern_chinese.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_nutrition.yaml
lm_eval/tasks/cmmlu/cmmlu_default_nutrition.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_philosophy.yaml
lm_eval/tasks/cmmlu/cmmlu_default_philosophy.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_professional_accounting.yaml
...al/tasks/cmmlu/cmmlu_default_professional_accounting.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_professional_law.yaml
lm_eval/tasks/cmmlu/cmmlu_default_professional_law.yaml
+4
-0
lm_eval/tasks/cmmlu/cmmlu_default_professional_medicine.yaml
lm_eval/tasks/cmmlu/cmmlu_default_professional_medicine.yaml
+4
-0
No files found.
lm_eval/tasks/cmmlu/cmmlu_default_high_school_geography.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
high_school_geography"
"
description"
:
"
以下是关于高中地理的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_high_school_geography"
lm_eval/tasks/cmmlu/cmmlu_default_high_school_mathematics.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
high_school_mathematics"
"
description"
:
"
以下是关于高中数学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_high_school_mathematics"
lm_eval/tasks/cmmlu/cmmlu_default_high_school_physics.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
high_school_physics"
"
description"
:
"
以下是关于高中物理学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_high_school_physics"
lm_eval/tasks/cmmlu/cmmlu_default_high_school_politics.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
high_school_politics"
"
description"
:
"
以下是关于高中政治的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_high_school_politics"
lm_eval/tasks/cmmlu/cmmlu_default_human_sexuality.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
human_sexuality"
"
description"
:
"
以下是关于人类性行为的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_human_sexuality"
lm_eval/tasks/cmmlu/cmmlu_default_international_law.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
international_law"
"
description"
:
"
以下是关于国际法学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_international_law"
lm_eval/tasks/cmmlu/cmmlu_default_journalism.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
journalism"
"
description"
:
"
以下是关于新闻学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_journalism"
lm_eval/tasks/cmmlu/cmmlu_default_jurisprudence.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
jurisprudence"
"
description"
:
"
以下是关于法理学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_jurisprudence"
lm_eval/tasks/cmmlu/cmmlu_default_legal_and_moral_basis.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
legal_and_moral_basis"
"
description"
:
"
以下是关于法律与道德基础的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_legal_and_moral_basis"
lm_eval/tasks/cmmlu/cmmlu_default_logical.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
logical"
"
description"
:
"
以下是关于逻辑学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_logical"
lm_eval/tasks/cmmlu/cmmlu_default_machine_learning.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
machine_learning"
"
description"
:
"
以下是关于机器学习的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_machine_learning"
lm_eval/tasks/cmmlu/cmmlu_default_management.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
management"
"
description"
:
"
以下是关于管理学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_management"
lm_eval/tasks/cmmlu/cmmlu_default_marketing.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
marketing"
"
description"
:
"
以下是关于市场营销的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_marketing"
lm_eval/tasks/cmmlu/cmmlu_default_marxist_theory.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
marxist_theory"
"
description"
:
"
以下是关于马克思主义理论的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_marxist_theory"
lm_eval/tasks/cmmlu/cmmlu_default_modern_chinese.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
modern_chinese"
"
description"
:
"
以下是关于现代汉语的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_modern_chinese"
lm_eval/tasks/cmmlu/cmmlu_default_nutrition.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
nutrition"
"
description"
:
"
以下是关于营养学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_nutrition"
lm_eval/tasks/cmmlu/cmmlu_default_philosophy.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
philosophy"
"
description"
:
"
以下是关于哲学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_philosophy"
lm_eval/tasks/cmmlu/cmmlu_default_professional_accounting.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
professional_accounting"
"
description"
:
"
以下是关于专业会计的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_professional_accounting"
lm_eval/tasks/cmmlu/cmmlu_default_professional_law.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
professional_law"
"
description"
:
"
以下是关于专业法学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_professional_law"
lm_eval/tasks/cmmlu/cmmlu_default_professional_medicine.yaml
0 → 100644
View file @
bdddfec2
"
dataset_name"
:
"
professional_medicine"
"
description"
:
"
以下是关于专业医学的单项选择题,请直接给出正确答案的选项。
\n\n
"
"
include"
:
"
_default_template_yaml"
"
task"
:
"
cmmlu_professional_medicine"
Prev
1
2
3
4
5
6
7
Next
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment