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
af6e46fd
Commit
af6e46fd
authored
Dec 08, 2023
by
lintangsutawika
Browse files
added prompt and output variations
parent
66421b57
Changes
32
Hide whitespace changes
Inline
Side-by-side
Showing
20 changed files
with
42 additions
and
81 deletions
+42
-81
lm_eval/tasks/mathqa/alternative_worlds/mathqa_alt.yaml
lm_eval/tasks/mathqa/alternative_worlds/mathqa_alt.yaml
+0
-10
lm_eval/tasks/mathqa/alternative_worlds/output_variation/README.md
...asks/mathqa/alternative_worlds/output_variation/README.md
+0
-2
lm_eval/tasks/mathqa/alternative_worlds/output_variation/_mathqa_alt_yaml
...thqa/alternative_worlds/output_variation/_mathqa_alt_yaml
+0
-5
lm_eval/tasks/mathqa/alternative_worlds/output_variation/mathqa_alt.yaml
...athqa/alternative_worlds/output_variation/mathqa_alt.yaml
+10
-0
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_01/a.yaml
...athqa/alternative_worlds/output_variation/style_01/a.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_01/b.yaml
...athqa/alternative_worlds/output_variation/style_01/b.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_01/c.yaml
...athqa/alternative_worlds/output_variation/style_01/c.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_02/a.yaml
...athqa/alternative_worlds/output_variation/style_02/a.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_02/b.yaml
...athqa/alternative_worlds/output_variation/style_02/b.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_02/c.yaml
...athqa/alternative_worlds/output_variation/style_02/c.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_03/a.yaml
...athqa/alternative_worlds/output_variation/style_03/a.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_03/b.yaml
...athqa/alternative_worlds/output_variation/style_03/b.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_03/c.yaml
...athqa/alternative_worlds/output_variation/style_03/c.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_04/a.yaml
...athqa/alternative_worlds/output_variation/style_04/a.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_04/b.yaml
...athqa/alternative_worlds/output_variation/style_04/b.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_04/c.yaml
...athqa/alternative_worlds/output_variation/style_04/c.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_05/a.yaml
...athqa/alternative_worlds/output_variation/style_05/a.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_05/b.yaml
...athqa/alternative_worlds/output_variation/style_05/b.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_05/c.yaml
...athqa/alternative_worlds/output_variation/style_05/c.yaml
+2
-4
lm_eval/tasks/mathqa/alternative_worlds/output_variation/style_06/a.yaml
...athqa/alternative_worlds/output_variation/style_06/a.yaml
+2
-4
No files found.
lm_eval/tasks/mathqa/alternative_worlds/mathqa_alt.yaml
deleted
100644 → 0
View file @
66421b57
group
:
mathqa_alt
task
:
-
mathqa_01
-
mathqa_02
-
mathqa_03
-
mathqa_04
-
mathqa_05
-
mathqa_06
-
mathqa_07
-
mathqa_08
lm_eval/tasks/mathqa/alternative_worlds/README.md
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
README.md
View file @
af6e46fd
Investigate affect of letter options
-
(A)
-
A)
...
...
@@ -9,7 +8,6 @@ Investigate affect of letter options
-
a)
-
a.
-
a
\t
Answer types:
-
letters only
-
original option
...
...
lm_eval/tasks/mathqa/alternative_worlds/_mathqa_alt_yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
_mathqa_alt_yaml
View file @
af6e46fd
...
...
@@ -6,10 +6,5 @@ test_split: test
doc_to_target: "{{['a', 'b', 'c', 'd', 'e'].index(correct)}}"
metric_list:
- metric: acc
aggregation: mean
higher_is_better: true
- metric: acc_norm
aggregation: mean
higher_is_better: true
- metric: brier_score
higher_is_better: false
lm_eval/tasks/mathqa/alternative_worlds/output_variation/mathqa_alt.yaml
0 → 100644
View file @
af6e46fd
group
:
mathqa_alt_ov
task
:
-
mathqa_alt_ov_01
-
mathqa_alt_ov_02
-
mathqa_alt_ov_03
-
mathqa_alt_ov_04
-
mathqa_alt_ov_05
-
mathqa_alt_ov_06
-
mathqa_alt_ov_07
-
mathqa_alt_ov_08
lm_eval/tasks/mathqa/alternative_worlds/style_01/a.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_01/a.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_01
group_alias
:
style_01
task
:
mathqa_01a
task_alias
:
a
group
:
mathqa_alt_ov_01
task
:
mathqa_alt_ov_01a
doc_to_text
:
!function
../styles.template_01
doc_to_choice
:
!function
../styles.choice_01a
lm_eval/tasks/mathqa/alternative_worlds/style_01/b.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_01/b.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_01
group_alias
:
style_01
task
:
mathqa_01b
task_alias
:
b
group
:
mathqa_alt_ov_01
task
:
mathqa_alt_ov_01b
doc_to_text
:
!function
../styles.template_01
doc_to_choice
:
!function
../styles.choice_01b
lm_eval/tasks/mathqa/alternative_worlds/style_01/c.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_01/c.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_01
group_alias
:
style_01
task
:
mathqa_01c
task_alias
:
c
group
:
mathqa_alt_ov_01
task
:
mathqa_alt_ov_01c
doc_to_text
:
!function
../styles.template_01
doc_to_choice
:
!function
../styles.choice_01c
lm_eval/tasks/mathqa/alternative_worlds/style_02/a.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_02/a.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_02
group_alias
:
style_02
task
:
mathqa_02a
task_alias
:
a
group
:
mathqa_alt_ov_02
task
:
mathqa_alt_ov_02a
doc_to_text
:
!function
../styles.template_02
doc_to_choice
:
!function
../styles.choice_02a
lm_eval/tasks/mathqa/alternative_worlds/style_02/b.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_02/b.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_02
group_alias
:
style_02
task
:
mathqa_02b
task_alias
:
b
group
:
mathqa_alt_ov_02
task
:
mathqa_alt_ov_02b
doc_to_text
:
!function
../styles.template_02
doc_to_choice
:
!function
../styles.choice_02b
lm_eval/tasks/mathqa/alternative_worlds/style_02/c.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_02/c.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_02
group_alias
:
style_02
task
:
mathqa_02c
task_alias
:
c
group
:
mathqa_alt_ov_02
task
:
mathqa_alt_ov_02c
doc_to_text
:
!function
../styles.template_02
doc_to_choice
:
!function
../styles.choice_02c
lm_eval/tasks/mathqa/alternative_worlds/style_03/a.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_03/a.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_03
group_alias
:
style_03
task
:
mathqa_03a
task_alias
:
a
group
:
mathqa_alt_ov_03
task
:
mathqa_alt_ov_03a
doc_to_text
:
!function
../styles.template_03
doc_to_choice
:
!function
../styles.choice_03a
lm_eval/tasks/mathqa/alternative_worlds/style_03/b.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_03/b.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_03
group_alias
:
style_03
task
:
mathqa_03b
task_alias
:
b
group
:
mathqa_alt_ov_03
task
:
mathqa_alt_ov_03b
doc_to_text
:
!function
../styles.template_03
doc_to_choice
:
!function
../styles.choice_03b
lm_eval/tasks/mathqa/alternative_worlds/style_03/c.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_03/c.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_03
group_alias
:
style_03
task
:
mathqa_03c
task_alias
:
c
group
:
mathqa_alt_ov_03
task
:
mathqa_alt_ov_03c
doc_to_text
:
!function
../styles.template_03
doc_to_choice
:
!function
../styles.choice_03c
lm_eval/tasks/mathqa/alternative_worlds/style_04/a.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_04/a.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_04
group_alias
:
style_04
task
:
mathqa_04a
task_alias
:
a
group
:
mathqa_alt_ov_04
task
:
mathqa_alt_ov_04a
doc_to_text
:
!function
../styles.template_04
doc_to_choice
:
!function
../styles.choice_04a
lm_eval/tasks/mathqa/alternative_worlds/style_04/b.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_04/b.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_04
group_alias
:
style_04
task
:
mathqa_04b
task_alias
:
b
group
:
mathqa_alt_ov_04
task
:
mathqa_alt_ov_04b
doc_to_text
:
!function
../styles.template_04
doc_to_choice
:
!function
../styles.choice_04b
lm_eval/tasks/mathqa/alternative_worlds/style_04/c.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_04/c.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_04
group_alias
:
style_04
task
:
mathqa_04c
task_alias
:
c
group
:
mathqa_alt_ov_04
task
:
mathqa_alt_ov_04c
doc_to_text
:
!function
../styles.template_04
doc_to_choice
:
!function
../styles.choice_04c
lm_eval/tasks/mathqa/alternative_worlds/style_05/a.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_05/a.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_05
group_alias
:
style_05
task
:
mathqa_05a
task_alias
:
a
group
:
mathqa_alt_ov_05
task
:
mathqa_alt_ov_05a
doc_to_text
:
!function
../styles.template_05
doc_to_choice
:
!function
../styles.choice_05a
lm_eval/tasks/mathqa/alternative_worlds/style_05/b.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_05/b.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_05
group_alias
:
style_05
task
:
mathqa_05b
task_alias
:
b
group
:
mathqa_alt_ov_05
task
:
mathqa_alt_ov_05b
doc_to_text
:
!function
../styles.template_05
doc_to_choice
:
!function
../styles.choice_05b
lm_eval/tasks/mathqa/alternative_worlds/style_05/c.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_05/c.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_05
group_alias
:
style_05
task
:
mathqa_05c
task_alias
:
c
group
:
mathqa_alt_ov_05
task
:
mathqa_alt_ov_05c
doc_to_text
:
!function
../styles.template_05
doc_to_choice
:
!function
../styles.choice_05c
lm_eval/tasks/mathqa/alternative_worlds/style_06/a.yaml
→
lm_eval/tasks/mathqa/alternative_worlds/
output_variation/
style_06/a.yaml
View file @
af6e46fd
include
:
../_mathqa_alt_yaml
group
:
mathqa_06
group_alias
:
style_06
task
:
mathqa_06a
task_alias
:
a
group
:
mathqa_alt_ov_06
task
:
mathqa_alt_ov_06a
doc_to_text
:
!function
../styles.template_06
doc_to_choice
:
!function
../styles.choice_06a
Prev
1
2
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