open-llm-leaderboard-old/details_vilm__Quyen-Mini-v0.1

该数据集是在模型vilm/Quyen-Mini-v0.1的评估运行期间自动创建的，用于在Open LLM Leaderboard上进行评估。数据集由63个配置组成，每个配置对应一个评估任务。数据集由1次运行生成，每次运行可以在每个配置的特定分割中找到，分割名称使用运行的时间戳命名。train分割始终指向最新的结果。此外，还有一个名为results的配置，存储了所有运行的聚合结果，并用于计算和显示在Open LLM Leaderboard上的聚合指标。

原始信息汇总

数据集概述

数据集名称

• 名称: Evaluation run of vilm/Quyen-Mini-v0.1

数据集来源

• 创建背景: 该数据集是在模型 vilm/Quyen-Mini-v0.1 在 Open LLM Leaderboard 上的评估运行期间自动创建的。

数据集结构

• 配置数量: 该数据集包含 63 个配置，每个配置对应一个评估任务。
• 数据来源: 数据集从 1 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。
• 结果汇总: 一个额外的配置 "results" 存储所有运行的汇总结果，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_vilm__Quyen-Mini-v0.1", "harness_winogrande_5", split="train")

最新结果

• 最新结果时间戳: 2024-02-17T16:04:03.826177
• 详细结果: 包含多个任务的准确率（acc）、标准化准确率（acc_norm）、标准误差（acc_stderr、acc_norm_stderr）等指标。

配置详情

• 配置名称: harness_arc_challenge_25

  • 数据文件:
    • 分割: 2024_02_17T16_04_03.826177
      • 路径: **/details_harness|arc:challenge|25_2024-02-17T16-04-03.826177.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2024-02-17T16-04-03.826177.parquet

• 配置名称: harness_gsm8k_5

  • 数据文件:
    • 分割: 2024_02_17T16_04_03.826177
      • 路径: **/details_harness|gsm8k|5_2024-02-17T16-04-03.826177.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2024-02-17T16-04-03.826177.parquet

• 配置名称: harness_hellaswag_10

  • 数据文件:
    • 分割: 2024_02_17T16_04_03.826177
      • 路径: **/details_harness|hellaswag|10_2024-02-17T16-04-03.826177.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2024-02-17T16-04-03.826177.parquet

• 配置名称: harness_hendrycksTest_5

  • 数据文件:
    • 分割: 2024_02_17T16_04_03.826177
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-international_law|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-management|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-marketing|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-professional_psychology|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-public_relations|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-security_studies|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-sociology|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-us_foreign_policy|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-virology|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-world_religions|5_2024-02-17T16-04-03.826177.parquet
    • 分割: latest
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-02-17T16-04-03.826177.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-02-17T1