nit

lm_eval/tasks/gsm8k/gsm8k-cot-llama.yaml

@@ -55,16 +55,9 @@ tag:
 metadata:
   version: 3.0
 metric_list:
-- aggregation: mean
-  higher_is_better: true
-  ignore_case: true
-  ignore_punctuation: false
-  metric: exact_match
-  regexes_to_ignore:
-  - ','
-  - \$
-  - '(?s).*#### '
-  - \.$
+  - metric: math_verify
+    aggregation: mean
+    higher_is_better: true
 num_fewshot: 8
 output_type: generate_until
 repeats: 1