Merge branch 'big-refactor' of...

Merge branch 'big-refactor' of https://github.com/EleutherAI/lm-evaluation-harness into openai_completions

lm_eval/tasks/bigbench/generate_until/metaphor_boolean.yaml

+# Generated by utils.py
+dataset_name: metaphor_boolean_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_metaphor_boolean_generate_until

lm_eval/tasks/bigbench/generate_until/metaphor_understanding.yaml

+# Generated by utils.py
+dataset_name: metaphor_understanding_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_metaphor_understanding_generate_until

lm_eval/tasks/bigbench/generate_until/minute_mysteries_qa.yaml

+# Generated by utils.py
+dataset_name: minute_mysteries_qa_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_minute_mysteries_qa_generate_until

lm_eval/tasks/bigbench/generate_until/misconceptions.yaml

+# Generated by utils.py
+dataset_name: misconceptions_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_misconceptions_generate_until

lm_eval/tasks/bigbench/generate_until/misconceptions_russian.yaml

+# Generated by utils.py
+dataset_name: misconceptions_russian_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_misconceptions_russian_generate_until

lm_eval/tasks/bigbench/generate_until/mnist_ascii.yaml

+# Generated by utils.py
+dataset_name: mnist_ascii_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_mnist_ascii_generate_until

lm_eval/tasks/bigbench/generate_until/modified_arithmetic.yaml

+# Generated by utils.py
+dataset_name: modified_arithmetic_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_modified_arithmetic_generate_until

lm_eval/tasks/bigbench/generate_until/moral_permissibility.yaml

+# Generated by utils.py
+dataset_name: moral_permissibility_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_moral_permissibility_generate_until

lm_eval/tasks/bigbench/generate_until/movie_dialog_same_or_different.yaml

+# Generated by utils.py
+dataset_name: movie_dialog_same_or_different_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_movie_dialog_same_or_different_generate_until

lm_eval/tasks/bigbench/generate_until/movie_recommendation.yaml

+# Generated by utils.py
+dataset_name: movie_recommendation_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_movie_recommendation_generate_until

lm_eval/tasks/bigbench/generate_until/mult_data_wrangling.yaml

+# Generated by utils.py
+dataset_name: mult_data_wrangling_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_mult_data_wrangling_generate_until

lm_eval/tasks/bigbench/generate_until/multiemo.yaml

+# Generated by utils.py
+dataset_name: multiemo_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_multiemo_generate_until

lm_eval/tasks/bigbench/generate_until/natural_instructions.yaml

+# Generated by utils.py
+dataset_name: natural_instructions_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_natural_instructions_generate_until

lm_eval/tasks/bigbench/generate_until/navigate.yaml

+# Generated by utils.py
+dataset_name: navigate_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_navigate_generate_until

lm_eval/tasks/bigbench/generate_until/nonsense_words_grammar.yaml

+# Generated by utils.py
+dataset_name: nonsense_words_grammar_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_nonsense_words_grammar_generate_until

lm_eval/tasks/bigbench/generate_until/novel_concepts.yaml

+# Generated by utils.py
+dataset_name: novel_concepts_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_novel_concepts_generate_until

lm_eval/tasks/bigbench/generate_until/object_counting.yaml

+# Generated by utils.py
+dataset_name: object_counting_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_object_counting_generate_until

lm_eval/tasks/bigbench/generate_until/odd_one_out.yaml

+# Generated by utils.py
+dataset_name: odd_one_out_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_odd_one_out_generate_until

lm_eval/tasks/bigbench/generate_until/operators.yaml

+# Generated by utils.py
+dataset_name: operators_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_operators_generate_until

lm_eval/tasks/bigbench/generate_until/paragraph_segmentation.yaml

+# Generated by utils.py
+dataset_name: paragraph_segmentation_zero_shot
+include: ../generate_until_template_yaml
+task: bigbench_paragraph_segmentation_generate_until