添加Megatron项目

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

+# Generated by utils.py
+dataset_name: cs_algorithms_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_cs_algorithms_multiple_choice

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

+# Generated by utils.py
+dataset_name: dark_humor_detection_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_dark_humor_detection_multiple_choice

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

+# Generated by utils.py
+dataset_name: date_understanding_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_date_understanding_multiple_choice

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

+# Generated by utils.py
+dataset_name: disambiguation_qa_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_disambiguation_qa_multiple_choice

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

+# Generated by utils.py
+dataset_name: discourse_marker_prediction_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_discourse_marker_prediction_multiple_choice

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

+# Generated by utils.py
+dataset_name: disfl_qa_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_disfl_qa_multiple_choice

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

+# Generated by utils.py
+dataset_name: dyck_languages_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_dyck_languages_multiple_choice

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

+# Generated by utils.py
+dataset_name: elementary_math_qa_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_elementary_math_qa_multiple_choice

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

+# Generated by utils.py
+dataset_name: emoji_movie_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_emoji_movie_multiple_choice

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

+# Generated by utils.py
+dataset_name: emojis_emotion_prediction_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_emojis_emotion_prediction_multiple_choice

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

+# Generated by utils.py
+dataset_name: empirical_judgments_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_empirical_judgments_multiple_choice

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

+# Generated by utils.py
+dataset_name: english_proverbs_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_english_proverbs_multiple_choice

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

+# Generated by utils.py
+dataset_name: english_russian_proverbs_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_english_russian_proverbs_multiple_choice

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

+# Generated by utils.py
+dataset_name: entailed_polarity_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_entailed_polarity_multiple_choice

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

+# Generated by utils.py
+dataset_name: entailed_polarity_hindi_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_entailed_polarity_hindi_multiple_choice

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

+# Generated by utils.py
+dataset_name: epistemic_reasoning_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_epistemic_reasoning_multiple_choice

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

+# Generated by utils.py
+dataset_name: evaluating_information_essentiality_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_evaluating_information_essentiality_multiple_choice

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

+# Generated by utils.py
+dataset_name: fact_checker_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_fact_checker_multiple_choice

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

+# Generated by utils.py
+dataset_name: fantasy_reasoning_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_fantasy_reasoning_multiple_choice

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

+# Generated by utils.py
+dataset_name: few_shot_nlg_zero_shot
+include: ../multiple_choice_template_yaml
+task: bigbench_few_shot_nlg_multiple_choice