five

open-llm-leaderboard-old/details_louisbrulenaudet__Pearl-3x7B

收藏
Hugging Face2024-02-09 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_louisbrulenaudet__Pearl-3x7B
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在对louisbrulenaudet/Pearl-3x7B模型进行评估运行期间自动创建的,用于Open LLM排行榜。数据集包含63个配置,每个配置对应一个评估任务。数据集来自1次运行,每次运行以运行的时间戳命名为特定的分割。train分割始终指向最新结果。额外的results配置存储了运行的所有聚合结果,用于在排行榜上计算和显示聚合指标。数据集结构允许使用HuggingFace数据集库加载运行的详细信息。

该数据集是在对louisbrulenaudet/Pearl-3x7B模型进行评估运行期间自动创建的,用于Open LLM排行榜。数据集包含63个配置,每个配置对应一个评估任务。数据集来自1次运行,每次运行以运行的时间戳命名为特定的分割。train分割始终指向最新结果。额外的results配置存储了运行的所有聚合结果,用于在排行榜上计算和显示聚合指标。数据集结构允许使用HuggingFace数据集库加载运行的详细信息。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集基本信息

  • 数据集名称: Evaluation run of louisbrulenaudet/Pearl-3x7B
  • 数据集来源: 自动创建于模型 louisbrulenaudet/Pearl-3x7BOpen LLM Leaderboard 的评估运行中。
  • 数据集组成: 包含 63 个配置,每个配置对应一个评估任务。
  • 数据集创建: 从 1 次运行中创建,每个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。"train" 分割始终指向最新的结果。
  • 额外配置: 一个名为 "results" 的配置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。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_louisbrulenaudet__Pearl-3x7B", "harness_winogrande_5", split="train")

最新结果

  • 最新结果来源: 2024-02-09T22:08:19.014926 运行
  • 结果示例: python { "all": { "acc": 0.6440565726482342, "acc_stderr": 0.031991772306209865, "acc_norm": 0.6465517429934861, "acc_norm_stderr": 0.032625459341892864, "mc1": 0.35495716034271724, "mc1_stderr": 0.016750862381375898, "mc2": 0.5216733642218897, "mc2_stderr": 0.015427199436320826 }, "harness|arc:challenge|25": { "acc": 0.6228668941979523, "acc_stderr": 0.014163366896192601, "acc_norm": 0.6552901023890785, "acc_norm_stderr": 0.01388881628678211 }, "harness|hellaswag|10": { "acc": 0.6701852220673172, "acc_stderr": 0.00469184866539907, "acc_norm": 0.8554072893845848, "acc_norm_stderr": 0.0035097096477918386 },

    更多任务结果...

}

配置信息

  • 配置名称: harness_arc_challenge_25

    • 数据文件:
      • 分割: 2024_02_09T22_08_19.014926
        • 路径: **/details_harness|arc:challenge|25_2024-02-09T22-08-19.014926.parquet
      • 分割: latest
        • 路径: **/details_harness|arc:challenge|25_2024-02-09T22-08-19.014926.parquet
  • 配置名称: harness_gsm8k_5

    • 数据文件:
      • 分割: 2024_02_09T22_08_19.014926
        • 路径: **/details_harness|gsm8k|5_2024-02-09T22-08-19.014926.parquet
      • 分割: latest
        • 路径: **/details_harness|gsm8k|5_2024-02-09T22-08-19.014926.parquet
  • 配置名称: harness_hellaswag_10

    • 数据文件:
      • 分割: 2024_02_09T22_08_19.014926
        • 路径: **/details_harness|hellaswag|10_2024-02-09T22-08-19.014926.parquet
      • 分割: latest
        • 路径: **/details_harness|hellaswag|10_2024-02-09T22-08-19.014926.parquet
  • 配置名称: harness_hendrycksTest_5

    • 数据文件:
      • 分割: 2024_02_09T22_08_19.014926
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-management|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-02-09T22-08-19.014926.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2024-02-09T22-08-19.014926.parquet
          • `**/details_harness|hendrycksTest-professional_medicine|5_2024
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