five

open-llm-leaderboard/details_The-Face-Of-Goonery__huginnv1.2

收藏
Hugging Face2023-10-15 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_The-Face-Of-Goonery__huginnv1.2
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型The-Face-Of-Goonery/huginnv1.2时自动创建的,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分割存储在配置中,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,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The-Face-Of-Goonery/huginnv1.2Open LLM Leaderboard上的评估运行期间自动创建的。

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。
  • 数据集由2次运行创建,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。
  • 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。

最新结果

以下是2023-10-15T03:33:52.049592运行的最新结果:

python { "all": { "em": 0.13464765100671142, "em_stderr": 0.0034957110748356193, "f1": 0.20755138422818709, "f1_stderr": 0.0036341951060626636, "acc": 0.421953322606337, "acc_stderr": 0.01004266408410234 }, "harness|drop|3": { "em": 0.13464765100671142, "em_stderr": 0.0034957110748356193, "f1": 0.20755138422818709, "f1_stderr": 0.0036341951060626636 }, "harness|gsm8k|5": { "acc": 0.09173616376042457, "acc_stderr": 0.00795094214833933 }, "harness|winogrande|5": { "acc": 0.7521704814522494, "acc_stderr": 0.01213438601986535 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割: 2023_08_09T23_01_31.106825
      • 路径: **/details_harness|arc:challenge|25_2023-08-09T23:01:31.106825.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2023-08-09T23:01:31.106825.parquet
  • harness_drop_3

    • 分割: 2023_10_15T03_33_52.049592
      • 路径: **/details_harness|drop|3_2023-10-15T03-33-52.049592.parquet
    • 分割: latest
      • 路径: **/details_harness|drop|3_2023-10-15T03-33-52.049592.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_15T03_33_52.049592
      • 路径: **/details_harness|gsm8k|5_2023-10-15T03-33-52.049592.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2023-10-15T03-33-52.049592.parquet
  • harness_hellaswag_10

    • 分割: 2023_08_09T23_01_31.106825
      • 路径: **/details_harness|hellaswag|10_2023-08-09T23:01:31.106825.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2023-08-09T23:01:31.106825.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_08_09T23_01_31.106825
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-08-09T23:01:31.106825.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-08-09T23:01:31.106825.parquet
        • `**/details_harness|hendrycksTest-professional_psychology|5_2023-08-0
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