add bigbench

lm_eval/tasks/bigbench/greedy_until/unit_conversion.yaml

+# Generated by utils.py
+dataset_name: unit_conversion
+include: ../greedy_until_template_yaml
+task: bigbench_unit_conversion_greedy_until

lm_eval/tasks/bigbench/greedy_until/unit_interpretation.yaml

+# Generated by utils.py
+dataset_name: unit_interpretation
+include: ../greedy_until_template_yaml
+task: bigbench_unit_interpretation_greedy_until

lm_eval/tasks/bigbench/greedy_until/unnatural_in_context_learning.yaml

+# Generated by utils.py
+dataset_name: unnatural_in_context_learning
+include: ../greedy_until_template_yaml
+task: bigbench_unnatural_in_context_learning_greedy_until

lm_eval/tasks/bigbench/greedy_until/vitaminc_fact_verification.yaml

+# Generated by utils.py
+dataset_name: vitaminc_fact_verification
+include: ../greedy_until_template_yaml
+task: bigbench_vitaminc_fact_verification_greedy_until

lm_eval/tasks/bigbench/greedy_until/what_is_the_tao.yaml

+# Generated by utils.py
+dataset_name: what_is_the_tao
+include: ../greedy_until_template_yaml
+task: bigbench_what_is_the_tao_greedy_until

lm_eval/tasks/bigbench/greedy_until/which_wiki_edit.yaml

+# Generated by utils.py
+dataset_name: which_wiki_edit
+include: ../greedy_until_template_yaml
+task: bigbench_which_wiki_edit_greedy_until

lm_eval/tasks/bigbench/greedy_until/winowhy.yaml

+# Generated by utils.py
+dataset_name: winowhy
+include: ../greedy_until_template_yaml
+task: bigbench_winowhy_greedy_until

lm_eval/tasks/bigbench/greedy_until/word_sorting.yaml

+# Generated by utils.py
+dataset_name: word_sorting
+include: ../greedy_until_template_yaml
+task: bigbench_word_sorting_greedy_until

lm_eval/tasks/bigbench/greedy_until/word_unscrambling.yaml

+# Generated by utils.py
+dataset_name: word_unscrambling
+include: ../greedy_until_template_yaml
+task: bigbench_word_unscrambling_greedy_until

lm_eval/tasks/bigbench/greedy_until_template_yaml

+group: bigbench
+dataset_path: bigbench
+output_type: greedy_until
+training_split: train
+validation_split: validation
+doc_to_text: inputs
+doc_to_target: "{{targets[0]}}"
+generation_kwargs:
+  max_length: 128
+metric_list:
+  - metric: exact_match
+    aggregation: mean
+    higher_is_better: true
+    ignore_punctuation: true

lm_eval/tasks/bigbench/multiple_choice/abstract_narrative_understanding.yaml

+# Generated by utils.py
+dataset_name: abstract_narrative_understanding
+include: ../multiple_choice_template_yaml
+task: bigbench_abstract_narrative_understanding_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/anachronisms.yaml

+# Generated by utils.py
+dataset_name: anachronisms
+include: ../multiple_choice_template_yaml
+task: bigbench_anachronisms_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/analogical_similarity.yaml

+# Generated by utils.py
+dataset_name: analogical_similarity
+include: ../multiple_choice_template_yaml
+task: bigbench_analogical_similarity_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/analytic_entailment.yaml

+# Generated by utils.py
+dataset_name: analytic_entailment
+include: ../multiple_choice_template_yaml
+task: bigbench_analytic_entailment_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/arithmetic.yaml

+# Generated by utils.py
+dataset_name: arithmetic
+include: ../multiple_choice_template_yaml
+task: bigbench_arithmetic_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/ascii_word_recognition.yaml

+# Generated by utils.py
+dataset_name: ascii_word_recognition
+include: ../multiple_choice_template_yaml
+task: bigbench_ascii_word_recognition_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/authorship_verification.yaml

+# Generated by utils.py
+dataset_name: authorship_verification
+include: ../multiple_choice_template_yaml
+task: bigbench_authorship_verification_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/auto_categorization.yaml

+# Generated by utils.py
+dataset_name: auto_categorization
+include: ../multiple_choice_template_yaml
+task: bigbench_auto_categorization_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/auto_debugging.yaml

+# Generated by utils.py
+dataset_name: auto_debugging
+include: ../multiple_choice_template_yaml
+task: bigbench_auto_debugging_multiple_choice

lm_eval/tasks/bigbench/multiple_choice/bbq_lite_json.yaml

+# Generated by utils.py
+dataset_name: bbq_lite_json
+include: ../multiple_choice_template_yaml
+task: bigbench_bbq_lite_json_multiple_choice