five

open-llm-leaderboard/details_AtAndDev__ShortKingv0.1

收藏
Hugging Face2023-10-24 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_AtAndDev__ShortKingv0.1
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型AtAndDev/ShortKingv0.1进行评估时自动生成的。数据集包含64个配置,每个配置对应一个评估任务。数据集由2次运行生成,每次运行的结果作为特定配置中的一个分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,还有一个名为results的配置,存储了所有运行的聚合结果,并用于在Open LLM Leaderboard上计算和显示聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

该数据集是在模型 AtAndDev/ShortKingv0.1Open LLM Leaderboard 上的评估运行期间自动创建的。数据集包含 64 个配置,每个配置对应一个评估任务。

数据集结构

数据集由 2 次运行创建,每次运行可以在每个配置中找到一个特定的分割,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_AtAndDev__ShortKingv0.1", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-24T00:55:05.543102 运行的最新结果

python { "all": { "em": 0.0050335570469798654, "em_stderr": 0.0007247385547751907, "f1": 0.054680159395973316, "f1_stderr": 0.0014128539394208607, "acc": 0.28246387417700025, "acc_stderr": 0.007901602410009655 }, "harness|drop|3": { "em": 0.0050335570469798654, "em_stderr": 0.0007247385547751907, "f1": 0.054680159395973316, "f1_stderr": 0.0014128539394208607 }, "harness|gsm8k|5": { "acc": 0.004548900682335102, "acc_stderr": 0.0018535550440036204 }, "harness|winogrande|5": { "acc": 0.5603788476716653, "acc_stderr": 0.01394964977601569 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割: 2023_10_03T17_59_37.972814
      • 路径: **/details_harness|arc:challenge|25_2023-10-03T17-59-37.972814.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2023-10-03T17-59-37.972814.parquet
  • harness_drop_3

    • 分割: 2023_10_24T00_55_05.543102
      • 路径: **/details_harness|drop|3_2023-10-24T00-55-05.543102.parquet
    • 分割: latest
      • 路径: **/details_harness|drop|3_2023-10-24T00-55-05.543102.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_24T00_55_05.543102
      • 路径: **/details_harness|gsm8k|5_2023-10-24T00-55-05.543102.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2023-10-24T00-55-05.543102.parquet
  • harness_hellaswag_10

    • 分割: 2023_10_03T17_59_37.972814
      • 路径: **/details_harness|hellaswag|10_2023-10-03T17-59-37.972814.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2023-10-03T17-59-37.972814.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_10_03T17_59_37.972814
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-management|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-10-03T17-59-37.972814.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-10-03T17-59-37.972814.parquet
        • `**/details_har
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作