open-llm-leaderboard/details_Gille__StrangeMerges_51-7B-dare_ties

该数据集是在模型Gille/StrangeMerges_51-7B-dare_ties在Open LLM Leaderboard上的评估运行期间自动创建的。数据集由63个配置组成，每个配置对应一个被评估的任务。数据集是从1次运行中创建的，每次运行作为每个配置中的一个特定分割，分割名称使用运行的时间戳。train分割始终指向最新的结果。一个额外的配置results存储了运行的所有聚合结果，用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了一个Python代码片段来加载运行的详细信息，并列出了特定运行的最新结果。

原始信息汇总

数据集概述

数据集名称

• pretty_name: Evaluation run of Gille/StrangeMerges_51-7B-dare_ties

数据集描述

• dataset_summary: 该数据集是在评估模型Gille/StrangeMerges_51-7B-dare_ties的过程中自动创建的，评估过程在Open LLM Leaderboard上进行。

数据集组成

• 配置数量: 63个
• 配置内容: 每个配置对应一个评估任务。
• 数据来源: 数据集由1次运行创建，每次运行在每个配置中作为一个特定的分割存在，分割名称使用运行的时间戳命名。
• 额外配置: 存在一个名为"results"的额外配置，用于存储所有运行的聚合结果，并用于计算和显示Open LLM Leaderboard上的聚合指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Gille__StrangeMerges_51-7B-dare_ties", "harness_winogrande_5", split="train")

最新结果

• 结果来源: 来自run 2024-04-02T22:47:20.672809。
• 结果内容: 包含多个任务的评估结果，如accuracy、accuracy_stderr等指标。

数据集配置详情

配置列表

• config_name: harness_arc_challenge_25

  • data_files:
    • split: 2024_04_02T22_47_20.672809
      • path: /details_harness|arc:challenge|25_2024-04-02T22-47-20.672809.parquet
    • split: latest
      • path: /details_harness|arc:challenge|25_2024-04-02T22-47-20.672809.parquet

• config_name: harness_gsm8k_5

  • data_files:
    • split: 2024_04_02T22_47_20.672809
      • path: /details_harness|gsm8k|5_2024-04-02T22-47-20.672809.parquet
    • split: latest
      • path: /details_harness|gsm8k|5_2024-04-02T22-47-20.672809.parquet

• config_name: harness_hellaswag_10

  • data_files:
    • split: 2024_04_02T22_47_20.672809
      • path: /details_harness|hellaswag|10_2024-04-02T22-47-20.672809.parquet
    • split: latest
      • path: /details_harness|hellaswag|10_2024-04-02T22-47-20.672809.parquet

• config_name: harness_hendrycksTest_5

  • data_files:
    • split: 2024_04_02T22_47_20.672809
      • path:
        • /details_harness|hendrycksTest-abstract_algebra|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-anatomy|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-astronomy|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-business_ethics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-clinical_knowledge|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-college_biology|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-college_chemistry|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-college_computer_science|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-college_mathematics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-college_medicine|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-college_physics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-computer_security|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-conceptual_physics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-econometrics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-electrical_engineering|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-elementary_mathematics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-formal_logic|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-global_facts|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_biology|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_chemistry|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_computer_science|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_european_history|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_geography|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_mathematics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_physics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_psychology|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_statistics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_us_history|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-high_school_world_history|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-human_aging|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-human_sexuality|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-international_law|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-jurisprudence|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-logical_fallacies|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-machine_learning|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-management|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-marketing|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-medical_genetics|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-miscellaneous|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-moral_disputes|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-moral_scenarios|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-nutrition|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-philosophy|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-prehistory|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-professional_accounting|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-professional_law|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-professional_medicine|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-professional_psychology|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-public_relations|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-security_studies|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-sociology|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-us_foreign_policy|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-virology|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-world_religions|5_2024-04-02T22-47-20.672809.parquet
    • split: latest
      • path:
        • /details_harness|hendrycksTest-abstract_algebra|5_2024-04-02T22-47-20.672809.parquet
        • /details_harness|hendrycksTest-anatomy|5_2024-04-02T22-47-20.672809.parquet
        • **/details_harness|hendrycksTest-astron