add many explicit group configs

lm_eval/tasks/agieval/lsat-rc.yaml

 include: aqua-rat.yaml
-group:
-  - agieval
-  - agieval_nous
-  - agieval_en
 task: agieval_lsat_rc
 dataset_path: hails/agieval-lsat-rc

lm_eval/tasks/agieval/math.yaml

-group:
-  - agieval
-  - agieval_en
 task: agieval_math
 dataset_path: hails/agieval-math
 dataset_name: null

lm_eval/tasks/agieval/sat-en-without-passage.yaml

 include: aqua-rat.yaml
-group:
-  - agieval
-  - agieval_nous
-  - agieval_en
 task: agieval_sat_en_without_passage
 dataset_path: hails/agieval-sat-en-without-passage

lm_eval/tasks/agieval/sat-en.yaml

 include: aqua-rat.yaml
-group:
-  - agieval
-  - agieval_nous
-  - agieval_en
 task: agieval_sat_en
 dataset_path: hails/agieval-sat-en

lm_eval/tasks/agieval/sat-math.yaml

 include: aqua-rat.yaml
-group:
-  - agieval
-  - agieval_nous
-  - agieval_en
 task: agieval_sat_math
 dataset_path: hails/agieval-sat-math

lm_eval/tasks/bbh/cot_fewshot/_bbh.yaml

@@ -27,7 +27,7 @@ task:
   - bbh_cot_fewshot_tracking_shuffled_objects_three_objects
   - bbh_cot_fewshot_web_of_lies
   - bbh_cot_fewshot_word_sorting
-aggregate_metric:
+aggregate_metric_list:
   - metric: exact_match
     aggregation: mean
     weight_by_size: true

lm_eval/tasks/bbh/cot_fewshot/_bbh_cot_fewshot.yaml

@@ -27,7 +27,7 @@ task:
   - bbh_cot_fewshot_tracking_shuffled_objects_three_objects
   - bbh_cot_fewshot_web_of_lies
   - bbh_cot_fewshot_word_sorting
-aggregate_metric:
+aggregate_metric_list:
   - metric: exact_match
     aggregation: mean
     weight_by_size: true

lm_eval/tasks/bbh/cot_zeroshot/_bbh_cot_zeroshot.yaml

@@ -27,7 +27,7 @@ task:
   - bbh_cot_zeroshot_tracking_shuffled_objects_three_objects
   - bbh_cot_zeroshot_web_of_lies
   - bbh_cot_zeroshot_word_sorting
-aggregate_metric:
+aggregate_metric_list:
   - metric: exact_match
     aggregation: mean
     weight_by_size: true

lm_eval/tasks/bbh/fewshot/_bbh_fewshot.yaml

@@ -27,7 +27,7 @@ task:
   - bbh_fewshot_tracking_shuffled_objects_three_objects
   - bbh_fewshot_web_of_lies
   - bbh_fewshot_word_sorting
-aggregate_metric:
+aggregate_metric_list:
   - metric: exact_match
     aggregation: mean
     weight_by_size: true

lm_eval/tasks/bbh/zeroshot/_bbh_zeroshot.yaml

@@ -27,7 +27,7 @@ task:
   - bbh_zeroshot_tracking_shuffled_objects_three_objects
   - bbh_zeroshot_web_of_lies
   - bbh_zeroshot_word_sorting
-aggregate_metric:
+aggregate_metric_list:
   - metric: exact_match
     aggregation: mean
     weight_by_size: true

lm_eval/tasks/storycloze/storycloze_2016.yaml

-group: storycloze
+tag: storycloze
 task: storycloze_2016
 dataset_path: story_cloze
 dataset_name: 2016

lm_eval/tasks/storycloze/storycloze_2018.yaml

-group: storycloze
+tag: storycloze
 task: storycloze_2018
 dataset_path: story_cloze
 dataset_name: 2018

lm_eval/tasks/super_glue/README.md

@@ -26,10 +26,14 @@ Homepage: https://super.gluebenchmark.com/
 }
 ```
 
-### Groups and Tasks
+### Groups, Tags, and Tasks
 
 #### Groups
 
+None.
+
+#### Tags
+
 * `super-glue-lm-eval-v1`: SuperGLUE eval adapted from LM Eval V1
 * `super-glue-t5-prompt`: SuperGLUE prompt and evaluation that matches the T5 paper (if using accelerate, will error if record is included.)