Merge remote-tracking branch 'origin/big-refactor' into big-refactor_dp

lm_eval/tasks/blimp/wh_questions_subject_gap_long_distance.yaml

 # Generated by utils.py
 dataset_name: wh_questions_subject_gap_long_distance
-include: template_yaml
+include: _template_yaml
 task: blimp_wh_questions_subject_gap_long_distance

lm_eval/tasks/blimp/wh_vs_that_no_gap.yaml

 # Generated by utils.py
 dataset_name: wh_vs_that_no_gap
-include: template_yaml
+include: _template_yaml
 task: blimp_wh_vs_that_no_gap

lm_eval/tasks/blimp/wh_vs_that_no_gap_long_distance.yaml

 # Generated by utils.py
 dataset_name: wh_vs_that_no_gap_long_distance
-include: template_yaml
+include: _template_yaml
 task: blimp_wh_vs_that_no_gap_long_distance

lm_eval/tasks/blimp/wh_vs_that_with_gap.yaml

 # Generated by utils.py
 dataset_name: wh_vs_that_with_gap
-include: template_yaml
+include: _template_yaml
 task: blimp_wh_vs_that_with_gap

lm_eval/tasks/blimp/wh_vs_that_with_gap_long_distance.yaml

 # Generated by utils.py
 dataset_name: wh_vs_that_with_gap_long_distance
-include: template_yaml
+include: _template_yaml
 task: blimp_wh_vs_that_with_gap_long_distance

lm_eval/tasks/ceval/_default_ceval_yaml

@@ -16,4 +16,4 @@ metric_list:
     aggregation: mean
     higher_is_better: true
 metadata:
-  version: "1.0"
+  - version: 1.0

lm_eval/tasks/cmmlu/_default_template_yaml

@@ -15,3 +15,5 @@ metric_list:
   - metric: acc_norm
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 0.0

lm_eval/tasks/code_x_glue/code-text/go.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: !function bleu.smoothed_bleu_4
     aggregation: mean
     higher_is_better: True
+metadata:
+  - version: 0.0

lm_eval/tasks/code_x_glue/code-text/java.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: !function bleu.smoothed_bleu_4
     aggregation: mean
     higher_is_better: True
+metadata:
+  - version: 0.0

lm_eval/tasks/code_x_glue/code-text/javascript.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: !function bleu.smoothed_bleu_4
     aggregation: mean
     higher_is_better: True
+metadata:
+  - version: 0.0

lm_eval/tasks/code_x_glue/code-text/php.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: !function bleu.smoothed_bleu_4
     aggregation: mean
     higher_is_better: True
+metadata:
+  - version: 0.0

lm_eval/tasks/code_x_glue/code-text/python.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: !function bleu.smoothed_bleu_4
     aggregation: mean
     higher_is_better: True
+metadata:
+  - version: 0.0

lm_eval/tasks/code_x_glue/code-text/ruby.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: !function bleu.smoothed_bleu_4
     aggregation: mean
     higher_is_better: True
+metadata:
+  - version: 2.0

lm_eval/tasks/coqa/default.yaml

@@ -18,3 +18,5 @@ metric_list:
   - metric: f1
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 2.0

lm_eval/tasks/crows_pairs/crows_pairs_english.yaml

@@ -19,3 +19,5 @@ metric_list:
   - metric: pct_stereotype
     aggregation: mean
     higher_is_better: false
+metadata:
+  - version: 1.0

lm_eval/tasks/csatqa/_default_csatqa_yaml

@@ -13,3 +13,5 @@ metric_list:
   - metric: acc_norm
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 0.0

lm_eval/tasks/drop/default.yaml

@@ -20,3 +20,5 @@ metric_list:
   - metric: f1
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 2.0

lm_eval/tasks/glue/cola/default.yaml

@@ -12,3 +12,5 @@ should_decontaminate: true
 doc_to_decontamination_query: sentence
 metric_list:
   - metric: mcc
+metadata:
+  - version: 1.0

lm_eval/tasks/glue/mnli/default.yaml

@@ -10,3 +10,5 @@ doc_to_target: label
 doc_to_choice: ["True", "Neither", "False"]
 metric_list:
   - metric: acc
+metadata:
+  - version: 1.0

lm_eval/tasks/glue/mrpc/default.yaml

@@ -11,3 +11,5 @@ doc_to_choice: ["no", "yes"]
 metric_list:
   - metric: acc
   - metric: f1
+metadata:
+  - version: 1.0