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
c6b5510c
Commit
c6b5510c
authored
Mar 08, 2024
by
LucWeber
Browse files
Add acknowledgements
parent
ef2c359f
Changes
4
Hide whitespace changes
Inline
Side-by-side
Showing
4 changed files
with
7 additions
and
0 deletions
+7
-0
lm_eval/tasks/tinyBenchmarks/README.md
lm_eval/tasks/tinyBenchmarks/README.md
+3
-0
lm_eval/tasks/tinyBenchmarks/utils_hellaswag.py
lm_eval/tasks/tinyBenchmarks/utils_hellaswag.py
+1
-0
lm_eval/tasks/tinyBenchmarks/utils_truthfulqa.py
lm_eval/tasks/tinyBenchmarks/utils_truthfulqa.py
+1
-0
lm_eval/tasks/tinyBenchmarks/utils_winogrande.py
lm_eval/tasks/tinyBenchmarks/utils_winogrande.py
+2
-0
No files found.
lm_eval/tasks/tinyBenchmarks/README.md
View file @
c6b5510c
...
...
@@ -10,6 +10,8 @@ The versatility of large language models (LLMs) led to the creation of diverse b
Homepage: -
All configs and utils mirror the ones from their original dataset!
### Groups and Tasks
#### Groups
...
...
@@ -95,6 +97,7 @@ Estimating performance for each scenario all at once
}
```
Please also reference the respective original dataset that you are using!
### Checklist
...
...
lm_eval/tasks/tinyBenchmarks/utils_hellaswag.py
View file @
c6b5510c
import
datasets
import
re
''' This code mirrors the utils of the original hellaswag task '''
def
preprocess
(
text
):
text
=
text
.
strip
()
...
...
lm_eval/tasks/tinyBenchmarks/utils_truthfulqa.py
View file @
c6b5510c
...
...
@@ -4,6 +4,7 @@ import numpy as np
from
rouge_score
import
rouge_scorer
,
scoring
''' This code mirrors the utils of the original truthful_qa task '''
def
process_results_mc2
(
doc
,
results
):
lls
,
is_greedy
=
zip
(
*
results
)
...
...
lm_eval/tasks/tinyBenchmarks/utils_winogrande.py
View file @
c6b5510c
''' This code mirrors the utils of the original winogrande task '''
def
doc_to_text
(
doc
):
answer_to_num
=
{
"1"
:
0
,
"2"
:
1
}
return
answer_to_num
[
doc
[
"answer"
]]
...
...
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