five

open-llm-leaderboard/details_meta-math__MetaMath-70B-V1.0

收藏
Hugging Face2023-10-27 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_meta-math__MetaMath-70B-V1.0
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型meta-math/MetaMath-70B-V1.0在Open LLM Leaderboard上的评估运行期间自动创建的。数据集由64个配置组成,每个配置对应一个被评估的任务。它包含来自2次运行的数据,每次运行都作为每个配置中的一个特定分割存储。train分割始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的指标。README还提供了如何使用Hugging Face datasets库加载数据集的示例。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 meta-math/MetaMath-70B-V1.0Open LLM Leaderboard 上的运行过程中自动创建的。

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。数据集从 2 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。

额外配置

一个额外的配置 "results" 存储了所有运行的聚合结果,用于计算并在 Open LLM Leaderboard 上显示聚合指标。

数据加载示例

以下是加载特定运行详细信息的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_meta-math__MetaMath-70B-V1.0", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-27T06:53:02.758124 运行 的最新结果: python { "all": { "em": 0.035968959731543626, "em_stderr": 0.0019069930004768872, "f1": 0.13366401006711418, "f1_stderr": 0.0024535730972056486, "acc": 0.6348774184360326, "acc_stderr": 0.01220774491883094 }, "harness|drop|3": { "em": 0.035968959731543626, "em_stderr": 0.0019069930004768872, "f1": 0.13366401006711418, "f1_stderr": 0.0024535730972056486 }, "harness|gsm8k|5": { "acc": 0.44655041698256254, "acc_stderr": 0.013693566549743144 }, "harness|winogrande|5": { "acc": 0.8232044198895028, "acc_stderr": 0.010721923287918735 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割:2023_10_04T06_01_20.870650
      • 路径:**/details_harness|arc:challenge|25_2023-10-04T06-01-20.870650.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-10-04T06-01-20.870650.parquet
  • harness_drop_3

    • 分割:2023_10_27T06_53_02.758124
      • 路径:**/details_harness|drop|3_2023-10-27T06-53-02.758124.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-27T06-53-02.758124.parquet
  • harness_gsm8k_5

    • 分割:2023_10_27T06_53_02.758124
      • 路径:**/details_harness|gsm8k|5_2023-10-27T06-53-02.758124.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-27T06-53-02.758124.parquet
  • harness_hellaswag_10

    • 分割:2023_10_04T06_01_20.870650
      • 路径:**/details_harness|hellaswag|10_2023-10-04T06-01-20.870650.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-10-04T06-01-20.870650.parquet
  • harness_hendrycksTest_5

    • 分割:2023_10_04T06_01_20.870650
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-management|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-10-04T06-01-20.870650.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-10-04T06-01-20.870650.parquet
        • `**/details_harness|hendrycksTest-professional_law|5_2023-10-04T06-01
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