添加Megatron项目

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/metaphor_understanding.yaml

+# Generated by utils.py
+dataset_name: metaphor_understanding_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_metaphor_understanding_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/minute_mysteries_qa.yaml

+# Generated by utils.py
+dataset_name: minute_mysteries_qa_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_minute_mysteries_qa_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/misconceptions.yaml

+# Generated by utils.py
+dataset_name: misconceptions_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_misconceptions_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/misconceptions_russian.yaml

+# Generated by utils.py
+dataset_name: misconceptions_russian_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_misconceptions_russian_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/mnist_ascii.yaml

+# Generated by utils.py
+dataset_name: mnist_ascii_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_mnist_ascii_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/modified_arithmetic.yaml

+# Generated by utils.py
+dataset_name: modified_arithmetic_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_modified_arithmetic_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/moral_permissibility.yaml

+# Generated by utils.py
+dataset_name: moral_permissibility_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_moral_permissibility_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/movie_dialog_same_or_different.yaml

+# Generated by utils.py
+dataset_name: movie_dialog_same_or_different_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_movie_dialog_same_or_different_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/movie_recommendation.yaml

+# Generated by utils.py
+dataset_name: movie_recommendation_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_movie_recommendation_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/mult_data_wrangling.yaml

+# Generated by utils.py
+dataset_name: mult_data_wrangling_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_mult_data_wrangling_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/multiemo.yaml

+# Generated by utils.py
+dataset_name: multiemo_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_multiemo_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/natural_instructions.yaml

+# Generated by utils.py
+dataset_name: natural_instructions_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_natural_instructions_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/navigate.yaml

+# Generated by utils.py
+dataset_name: navigate_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_navigate_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/nonsense_words_grammar.yaml

+# Generated by utils.py
+dataset_name: nonsense_words_grammar_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_nonsense_words_grammar_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/novel_concepts.yaml

+# Generated by utils.py
+dataset_name: novel_concepts_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_novel_concepts_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/object_counting.yaml

+# Generated by utils.py
+dataset_name: object_counting_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_object_counting_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/odd_one_out.yaml

+# Generated by utils.py
+dataset_name: odd_one_out_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_odd_one_out_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/operators.yaml

+# Generated by utils.py
+dataset_name: operators_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_operators_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/paragraph_segmentation.yaml

+# Generated by utils.py
+dataset_name: paragraph_segmentation_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_paragraph_segmentation_multiple_choice

LM-Evaluation-Harness-240310/lm_eval/tasks/bigbench/multiple_choice/parsinlu_qa.yaml

+# Generated by utils.py
+dataset_name: parsinlu_qa_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_parsinlu_qa_multiple_choice