five

open-llm-leaderboard/details_adept__persimmon-8b-chat

收藏
Hugging Face2023-10-11 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_adept__persimmon-8b-chat
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adept/persimmon-8b-chat进行评估时自动创建的。数据集由61个配置组成,每个配置对应一个评估任务。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。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Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

该数据集是在对模型 adept/persimmon-8b-chat 进行评估运行期间自动创建的。数据集包含61个配置,每个配置对应一个评估任务。数据集来源于1次运行,每个运行的详细结果可以在每个配置中找到,以运行的时间戳命名的特定分片形式存储。"train" 分片始终指向最新的结果。

数据集结构

数据集包含以下主要配置:

  • results: 存储所有运行结果的聚合数据,用于计算和显示在 Open LLM Leaderboard 上的聚合指标。

配置详情

  • harness_arc_challenge_25:

    • 分片: 2023_10_11T16_42_26.599502
    • 路径: **/details_harness|arc:challenge|25_2023-10-11T16-42-26.599502.parquet
    • 分片: latest
    • 路径: **/details_harness|arc:challenge|25_2023-10-11T16-42-26.599502.parquet
  • harness_hellaswag_10:

    • 分片: 2023_10_11T16_42_26.599502
    • 路径: **/details_harness|hellaswag|10_2023-10-11T16-42-26.599502.parquet
    • 分片: latest
    • 路径: **/details_harness|hellaswag|10_2023-10-11T16-42-26.599502.parquet
  • harness_hendrycksTest_5:

    • 分片: 2023_10_11T16_42_26.599502
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-11T16-42-26.599502.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-10-11T16-42-26.599502.parquet
      • `**/details_harness|hendrycksTest-high_school_governm
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