Spaces:

hallucinations-leaderboard
/

leaderboard

Running on CPU Upgrade

pminervini commited on Jan 19

Commit

be7e092

•

1 Parent(s): 136dad6

update

Files changed (1) hide show

src/backend/envs.py CHANGED Viewed

@@ -24,7 +24,7 @@ class Tasks(Enum):
     task1 = Task("triviaqa", "em", "TriviaQA", 64)  # 64, as in the ATLAS paper
     # TruthfulQA is intended as a zero-shot benchmark [5, 47]. https://owainevans.github.io/pdfs/truthfulQA_lin_evans.pdf
-    # task2 = Task("truthfulqa_gen", "rougeL_acc", "TruthfulQA Gen", 0)
     task3 = Task("truthfulqa_mc1", "acc", "TruthfulQA MC1", 0)
     task4 = Task("truthfulqa_mc2", "acc", "TruthfulQA MC2", 0)

     task1 = Task("triviaqa", "em", "TriviaQA", 64)  # 64, as in the ATLAS paper
     # TruthfulQA is intended as a zero-shot benchmark [5, 47]. https://owainevans.github.io/pdfs/truthfulQA_lin_evans.pdf
+    task2 = Task("truthfulqa_gen", "rougeL_acc", "TruthfulQA Gen", 0)
     task3 = Task("truthfulqa_mc1", "acc", "TruthfulQA MC1", 0)
     task4 = Task("truthfulqa_mc2", "acc", "TruthfulQA MC2", 0)