merged latest and added altworld files

lm_eval/tasks/hendrycks_ethics/utilitarianism_original_yaml

@@ -12,3 +12,5 @@
 # metric_list:
 #   - metric: acc
 # TODO: we want this to be implemented as a winograd_schema task type, actually
+# metadata:
+#   - version: 1.0

lm_eval/tasks/hendrycks_ethics/virtue.yaml

@@ -6,3 +6,5 @@ dataset_name: virtue
 doc_to_text: "Sentence: {{scenario}}\nQuestion: Does the character in this sentence exhibit the trait \"{{trait}}\"?\nAnswer:"
 doc_to_target: label
 doc_to_choice: ['no', 'yes']
+metadata:
+  - version: 1.0

lm_eval/tasks/lambada/lambada_openai.yaml

@@ -16,3 +16,5 @@ metric_list:
   - metric: acc
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 1.0

lm_eval/tasks/lambada/lambada_standard.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: acc
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 1.0

lm_eval/tasks/lambada_cloze/lambada_openai_cloze.yaml

@@ -16,3 +16,5 @@ metric_list:
   - metric: acc
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 1.0

lm_eval/tasks/lambada_cloze/lambada_standard_cloze.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: acc
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 1.0

lm_eval/tasks/lambada_multilingual/lambada_mt_en.yaml

@@ -16,3 +16,5 @@ metric_list:
   - metric: acc
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 1.0

lm_eval/tasks/logiqa/logiqa.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: acc_norm
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 1.0

lm_eval/tasks/logiqa2/logieval.yaml

@@ -23,3 +23,5 @@ filter_list:
         # https://github.com/openai/evals/blob/305b237cdb3884c7ddb6a5d12cb184a83551fcba/evals/api.py#L84
         regex_pattern: "^\\s*([A-D])"
       - function: "take_first"
+metadata:
+  - version: 0.0

lm_eval/tasks/logiqa2/logiqa2.yaml

@@ -17,3 +17,5 @@ metric_list:
   - metric: acc_norm
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 0.0

lm_eval/tasks/mathqa/alternative_worlds/_mathqa_alt_yaml

@@ -8,5 +8,8 @@ metric_list:
   - metric: acc
     aggregation: mean
     higher_is_better: true
+  - metric: acc_norm
+    aggregation: mean
+    higher_is_better: true
   - metric: brier_score
     higher_is_better: false

lm_eval/tasks/mathqa/mathqa.yaml

@@ -18,3 +18,5 @@ metric_list:
   - metric: acc_norm
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 1.0

lm_eval/tasks/mc_taco/default.yaml

@@ -11,3 +11,5 @@ doc_to_decontamination_query: "{{question}} {{sentence}}"
 metric_list:
   - metric: acc
   - metric: f1
+metadata:
+  - version: 1.0

lm_eval/tasks/mgsm/direct/direct_yaml

@@ -25,3 +25,5 @@ metric_list:
     higher_is_better: true
     ignore_case: true
     ignore_punctuation: true
+metadata:
+  - version: 0.0

lm_eval/tasks/mgsm/en_cot/cot_yaml

@@ -27,3 +27,5 @@ filter_list:
       - function: "regex"
         regex_pattern: "The answer is (\\-?[0-9\\.\\,]+)"
       - function: "take_first"
+metadata:
+  - version: 0.0

lm_eval/tasks/mgsm/native_cot/cot_yaml

@@ -27,3 +27,5 @@ filter_list:
       - function: "regex"
         regex_pattern: "The answer is (\\-?[0-9\\.\\,]+)"
       - function: "take_first"
+metadata:
+  - version: 1.0

lm_eval/tasks/minerva_math/minerva_math_algebra.yaml

@@ -19,3 +19,5 @@ metric_list:
   - metric: exact_match
     aggregation: mean
     higher_is_better: true
+metadata:
+  - version: 0.0

lm_eval/tasks/mmlu/default/_default_template_yaml

@@ -11,6 +11,5 @@ metric_list:
   - metric: acc
     aggregation: mean
     higher_is_better: true
-  - metric: brier_score
-    aggregation: mean
-    higher_is_better: false
+metadata:
+  - version: 0.0

lm_eval/tasks/mmlu/flan_cot_fewshot/_mmlu_flan_cot_fewshot_template_yaml

@@ -21,3 +21,5 @@ metric_list:
     higher_is_better: true
     ignore_case: true
     ignore_punctuation: true
+metadata:
+  - version: 0.0