open-llm-leaderboard/details_migtissera__Synthia-70B-v1.2
收藏数据集概述
该数据集是在评估模型 migtissera/Synthia-70B-v1.2 在 Open LLM Leaderboard 上的自动创建的。
数据集组成
- 数据集包含 64 个配置,每个配置对应一个评估任务。
- 数据集从 2 次运行中创建,每次运行可以在每个配置中找到特定的分割,分割名称使用运行的时间戳。
- "train" 分割始终指向最新的结果。
额外配置
- 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示 Open LLM Leaderboard 上的聚合指标。
加载数据示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_migtissera__Synthia-70B-v1.2", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-17T14:51:20.480254 运行的最新结果: python { "all": { "em": 0.364618288590604, "em_stderr": 0.004929197624393639, "f1": 0.42417365771812215, "f1_stderr": 0.004776577842624861, "acc": 0.5759284047791582, "acc_stderr": 0.011665477241539865 }, "harness|drop|3": { "em": 0.364618288590604, "em_stderr": 0.004929197624393639, "f1": 0.42417365771812215, "f1_stderr": 0.004776577842624861 }, "harness|gsm8k|5": { "acc": 0.3191811978771797, "acc_stderr": 0.012840345676251651 }, "harness|winogrande|5": { "acc": 0.8326756116811366, "acc_stderr": 0.010490608806828079 } }
配置详情
配置列表
harness_arc_challenge_25harness_drop_3harness_gsm8k_5harness_hellaswag_10harness_hendrycksTest_5harness_hendrycksTest_abstract_algebra_5harness_hendrycksTest_anatomy_5harness_hendrycksTest_astronomy_5harness_hendrycksTest_business_ethics_5harness_hendrycksTest_clinical_knowledge_5harness_hendrycksTest_college_biology_5harness_hendrycksTest_college_chemistry_5harness_hendrycksTest_college_computer_science_5harness_hendrycksTest_college_mathematics_5harness_hendrycksTest_college_medicine_5harness_hendrycksTest_college_physics_5harness_hendrycksTest_computer_security_5harness_hendrycksTest_conceptual_physics_5harness_hendrycksTest_econometrics_5harness_hendrycksTest_electrical_engineering_5harness_hendrycksTest_elementary_mathematics_5harness_hendrycksTest_formal_logic_5harness_hendrycksTest_global_facts_5harness_hendrycksTest_high_school_biology_5harness_hendrycksTest_high_school_chemistry_5harness_hendrycksTest_high_school_computer_science_5harness_hendrycksTest_high_school_european_history_5harness_hendrycksTest_high_school_geography_5harness_hendrycksTest_high_school_government_and_politics_5harness_hendrycksTest_high_school_macroeconomics_5harness_hendrycksTest_high_school_mathematics_5harness_hendrycksTest_high_school_microeconomics_5harness_hendrycksTest_high_school_physics_5harness_hendrycksTest_high_school_psychology_5harness_hendrycksTest_high_school_statistics_5harness_hendrycksTest_high_school_us_history_5harness_hendrycksTest_high_school_world_history_5harness_hendrycksTest_human_aging_5harness_hendrycksTest_human_sexuality_5harness_hendrycksTest_international_law_5harness_hendrycksTest_jurisprudence_5harness_hendrycksTest_logical_fallacies_5harness_hendrycksTest_machine_learning_5harness_hendrycksTest_management_5harness_hendrycksTest_marketing_5harness_hendrycksTest_medical_genetics_5harness_hendrycksTest_miscellaneous_5harness_hendrycksTest_moral_disputes_5harness_hendrycksTest_moral_scenarios_5harness_hendrycksTest_nutrition_5harness_hendrycksTest_philosophy_5harness_hendrycksTest_prehistory_5harness_hendrycksTest_professional_accounting_5harness_hendrycksTest_professional_law_5harness_hendrycksTest_professional_medicine_5harness_hendrycksTest_professional_psychology_5harness_hendrycksTest_public_relations_5harness_hendrycksTest_security_studies_5harness_hendrycksTest_sociology_5harness_hendrycksTest_us_foreign_policy_5harness_hendrycksTest_virology_5harness_hendrycksTest_world_religions_5
数据文件路径
每个配置包含多个数据文件路径,例如:
harness_arc_challenge_25包含details_harness|arc:challenge|25_2023-09-02T17:59:05.420313.parquetharness_drop_3包含details_harness|drop|3_2023-10-17T14-51-20.480254.parquetharness_gsm8k_5包含details_harness|gsm8k|5_2023-10-17T14-51-20.480254.parquetharness_hellaswag_10包含details_harness|hellaswag|10_2023-09-02T17:59:05.420313.parquetharness_hendrycksTest_5包含多个文件路径,如details_harness|hendrycksTest-abstract_algebra|5_2023-09-02T17:59:05.420313.parquet等。
每个配置的数据文件路径分为两个分割:2023_09_02T17_59_05.420313 和 latest。



