five

open-llm-leaderboard-old/details_Fredithefish__MadMix-v0.2

收藏
Hugging Face2023-12-13 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Fredithefish__MadMix-v0.2
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型Fredithefish/MadMix-v0.2进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个被评估的任务。数据集是从1次运行中创建的,每次运行在每个配置中作为一个特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,还有一个名为results的配置,存储了所有运行的聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。文件还提供了一个Python代码片段,用于加载运行中的详细信息,并列出了特定运行的最新结果。

该数据集是在Open LLM Leaderboard上对模型Fredithefish/MadMix-v0.2进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个被评估的任务。数据集是从1次运行中创建的,每次运行在每个配置中作为一个特定的分割,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,还有一个名为results的配置,存储了所有运行的聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。文件还提供了一个Python代码片段,用于加载运行中的详细信息,并列出了特定运行的最新结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

该数据集是在对模型 Fredithefish/MadMix-v0.2 进行评估运行时自动创建的,评估结果发布在 Open LLM Leaderboard 上。

数据集组成

  • 数据集包含 63 个配置,每个配置对应一个评估任务。
  • 数据集从 1 次运行中创建,每个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,用于计算和显示在 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_Fredithefish__MadMix-v0.2", "harness_winogrande_5", split="train")

最新结果

以下是 2023-12-13T14:05:21.026103 运行的最新结果

python { "all": { "acc": 0.6369931523249152, "acc_stderr": 0.032180405415322286, "acc_norm": 0.643635199288352, "acc_norm_stderr": 0.03283580466491587, "mc1": 0.386780905752754, "mc1_stderr": 0.017048857010515107, "mc2": 0.5578972226689095, "mc2_stderr": 0.015808791087791672 }, "harness|arc:challenge|25": { "acc": 0.6271331058020477, "acc_stderr": 0.014131176760131172, "acc_norm": 0.6484641638225256, "acc_norm_stderr": 0.013952413699600931 }, "harness|hellaswag|10": { "acc": 0.6575383389762995, "acc_stderr": 0.004735632975072384, "acc_norm": 0.8353913563035252, "acc_norm_stderr": 0.003700690995600887 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.33, "acc_stderr": 0.04725815626252605, "acc_norm": 0.33, "acc_norm_stderr": 0.04725815626252605 }, ... }

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2023_12_13T14_05_21.026103
      • path: **/details_harness|arc:challenge|25_2023-12-13T14-05-21.026103.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-12-13T14-05-21.026103.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_12_13T14_05_21.026103
      • path: **/details_harness|gsm8k|5_2023-12-13T14-05-21.026103.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-12-13T14-05-21.026103.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_12_13T14_05_21.026103
      • path: **/details_harness|hellaswag|10_2023-12-13T14-05-21.026103.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-12-13T14-05-21.026103.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_12_13T14_05_21.026103
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-12-13T14-05-21.026103.parquet ...
    • split: latest
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-12-13T14-05-21.026103.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-12-13T14-05-21.026103.parquet ... }
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