Skip to content

GitLab

Unverified Commit 054ed37f authored by Hailey Schoelkopf's avatar Hailey Schoelkopf Committed by GitHub
Browse files

Merge pull request #766 from EleutherAI/patch-crowspairs 

[Refactor] Patch crowspairs higher_is_better
parents 4f27f0b9 bede79c2
Show whitespace changes
Inline Side-by-side
Showing with 3 additions and 3 deletions
lm_eval/tasks/crows_pairs/crows_pairs_english.yaml
View file @ 054ed37f
...@@ -16,6 +16,6 @@ metric_list: ...@@ -16,6 +16,6 @@ metric_list:
- metric: likelihood_diff - metric: likelihood_diff
aggregation: mean aggregation: mean
higher_is_better: false higher_is_better: false
- metric: acc - metric: pct_stereotype
aggregation: mean aggregation: mean
higher_is_better: true higher_is_better: false
lm_eval/tasks/crows_pairs/utils.py
View file @ 054ed37f
...@@ -13,7 +13,7 @@ def process_results(doc, results): ...@@ -13,7 +13,7 @@ def process_results(doc, results):
# then treat this as predicting stereotyped sentence # then treat this as predicting stereotyped sentence
acc = 1.0 if likelihood1 > likelihood2 else 0.0 acc = 1.0 if likelihood1 > likelihood2 else 0.0
return {"likelihood_diff": diff, "acc": acc} return {"likelihood_diff": diff, "pct_stereotype": acc}
def doc_to_choice(doc): def doc_to_choice(doc):
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment