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
4277840b
Commit
4277840b
authored
Dec 27, 2023
by
lintangsutawika
Browse files
add piqa
parent
f38c7469
Changes
32
Hide whitespace changes
Inline
Side-by-side
Showing
20 changed files
with
156 additions
and
0 deletions
+156
-0
lm_eval/tasks/piqa/alternative_worlds/README.md
lm_eval/tasks/piqa/alternative_worlds/README.md
+20
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/_piqa_alt_ov_yaml
...iqa/alternative_worlds/output_variation/_piqa_alt_ov_yaml
+24
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/piqa_alt_ov.yaml
...piqa/alternative_worlds/output_variation/piqa_alt_ov.yaml
+10
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_01/a.yaml
.../piqa/alternative_worlds/output_variation/style_01/a.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_01/b.yaml
.../piqa/alternative_worlds/output_variation/style_01/b.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_01/c.yaml
.../piqa/alternative_worlds/output_variation/style_01/c.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_02/a.yaml
.../piqa/alternative_worlds/output_variation/style_02/a.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_02/b.yaml
.../piqa/alternative_worlds/output_variation/style_02/b.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_02/c.yaml
.../piqa/alternative_worlds/output_variation/style_02/c.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_03/a.yaml
.../piqa/alternative_worlds/output_variation/style_03/a.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_03/b.yaml
.../piqa/alternative_worlds/output_variation/style_03/b.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_03/c.yaml
.../piqa/alternative_worlds/output_variation/style_03/c.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_04/a.yaml
.../piqa/alternative_worlds/output_variation/style_04/a.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_04/b.yaml
.../piqa/alternative_worlds/output_variation/style_04/b.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_04/c.yaml
.../piqa/alternative_worlds/output_variation/style_04/c.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_05/a.yaml
.../piqa/alternative_worlds/output_variation/style_05/a.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_05/b.yaml
.../piqa/alternative_worlds/output_variation/style_05/b.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_05/c.yaml
.../piqa/alternative_worlds/output_variation/style_05/c.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_06/a.yaml
.../piqa/alternative_worlds/output_variation/style_06/a.yaml
+6
-0
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_06/b.yaml
.../piqa/alternative_worlds/output_variation/style_06/b.yaml
+6
-0
No files found.
lm_eval/tasks/piqa/alternative_worlds/README.md
0 → 100644
View file @
4277840b
Investigate affect of letter options
-
(A)
-
A)
-
A.
-
A
\t
-
(a)
-
a)
-
a.
-
a
\t
Answer types:
-
letters only
-
original option
-
just letter
-
letters + continuation
-
original option
-
just letter
-
continuation
lm_eval/tasks/piqa/alternative_worlds/output_variation/_piqa_alt_ov_yaml
0 → 100644
View file @
4277840b
group:
- ai2_arc
task: piqa
dataset_path: ai2_arc
dataset_name: ARC-Easy
output_type: multiple_choice
training_split: train
validation_split: validation
test_split: test
doc_to_text: "Question: {{question}}\nAnswer:"
doc_to_target: "{{choices.label.index(answerKey)}}"
doc_to_choice: "{{choices.text}}"
should_decontaminate: true
doc_to_decontamination_query: "Question: {{question}}\nAnswer:"
metric_list:
- metric: acc
aggregation: mean
higher_is_better: true
- metric: acc_norm
aggregation: mean
higher_is_better: true
- metric: brier_score
aggregation: brier_score
higher_is_better: false
lm_eval/tasks/piqa/alternative_worlds/output_variation/piqa_alt_ov.yaml
0 → 100644
View file @
4277840b
group
:
piqa_alt_ov
task
:
-
piqa_alt_ov_01
-
piqa_alt_ov_02
-
piqa_alt_ov_03
-
piqa_alt_ov_04
-
piqa_alt_ov_05
-
piqa_alt_ov_06
-
piqa_alt_ov_07
-
piqa_alt_ov_08
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_01/a.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_01
task
:
piqa_alt_ov_01a
doc_to_text
:
!function
../styles.template_01
doc_to_choice
:
!function
../styles.choice_01a
doc_to_decontamination_query
:
!function
../styles.template_01
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_01/b.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_01
task
:
piqa_alt_ov_01b
doc_to_text
:
!function
../styles.template_01
doc_to_choice
:
!function
../styles.choice_01b
doc_to_decontamination_query
:
!function
../styles.template_01
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_01/c.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_01
task
:
piqa_alt_ov_01c
doc_to_text
:
!function
../styles.template_01
doc_to_choice
:
!function
../styles.choice_01c
doc_to_decontamination_query
:
!function
../styles.template_01
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_02/a.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_02
task
:
piqa_alt_ov_02a
doc_to_text
:
!function
../styles.template_02
doc_to_choice
:
!function
../styles.choice_02a
doc_to_decontamination_query
:
!function
../styles.template_02
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_02/b.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_02
task
:
piqa_alt_ov_02b
doc_to_text
:
!function
../styles.template_02
doc_to_choice
:
!function
../styles.choice_02b
doc_to_decontamination_query
:
!function
../styles.template_02
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_02/c.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_02
task
:
piqa_alt_ov_02c
doc_to_text
:
!function
../styles.template_02
doc_to_choice
:
!function
../styles.choice_02c
doc_to_decontamination_query
:
!function
../styles.template_02
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_03/a.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_03
task
:
piqa_alt_ov_03a
doc_to_text
:
!function
../styles.template_03
doc_to_choice
:
!function
../styles.choice_03a
doc_to_decontamination_query
:
!function
../styles.template_03
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_03/b.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_03
task
:
piqa_alt_ov_03b
doc_to_text
:
!function
../styles.template_03
doc_to_choice
:
!function
../styles.choice_03b
doc_to_decontamination_query
:
!function
../styles.template_03
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_03/c.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_03
task
:
piqa_alt_ov_03c
doc_to_text
:
!function
../styles.template_03
doc_to_choice
:
!function
../styles.choice_03c
doc_to_decontamination_query
:
!function
../styles.template_03
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_04/a.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_04
task
:
piqa_alt_ov_04a
doc_to_text
:
!function
../styles.template_04
doc_to_choice
:
!function
../styles.choice_04a
doc_to_decontamination_query
:
!function
../styles.template_04
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_04/b.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_04
task
:
piqa_alt_ov_04b
doc_to_text
:
!function
../styles.template_04
doc_to_choice
:
!function
../styles.choice_04b
doc_to_decontamination_query
:
!function
../styles.template_04
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_04/c.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_04
task
:
piqa_alt_ov_04c
doc_to_text
:
!function
../styles.template_04
doc_to_choice
:
!function
../styles.choice_04c
doc_to_decontamination_query
:
!function
../styles.template_04
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_05/a.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_05
task
:
piqa_alt_ov_05a
doc_to_text
:
!function
../styles.template_05
doc_to_choice
:
!function
../styles.choice_05a
doc_to_decontamination_query
:
!function
../styles.template_05
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_05/b.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_05
task
:
piqa_alt_ov_05b
doc_to_text
:
!function
../styles.template_05
doc_to_choice
:
!function
../styles.choice_05b
doc_to_decontamination_query
:
!function
../styles.template_05
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_05/c.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_05
task
:
piqa_alt_ov_05c
doc_to_text
:
!function
../styles.template_05
doc_to_choice
:
!function
../styles.choice_05c
doc_to_decontamination_query
:
!function
../styles.template_05
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_06/a.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_06
task
:
piqa_alt_ov_06a
doc_to_text
:
!function
../styles.template_06
doc_to_choice
:
!function
../styles.choice_06a
doc_to_decontamination_query
:
!function
../styles.template_06
lm_eval/tasks/piqa/alternative_worlds/output_variation/style_06/b.yaml
0 → 100644
View file @
4277840b
include
:
../_piqa_alt_ov_yaml
group
:
piqa_alt_ov_06
task
:
piqa_alt_ov_06b
doc_to_text
:
!function
../styles.template_06
doc_to_choice
:
!function
../styles.choice_06b
doc_to_decontamination_query
:
!function
../styles.template_06
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