open-llm-leaderboard-old/details_Azazelle__Mocha-Dare-7b-ex

该数据集是在Open LLM Leaderboard上对模型Azazelle/Mocha-Dare-7b-ex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，results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。

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 Azazelle/Mocha-Dare-7b-ex
• 数据集来源: 自动创建于模型 Azazelle/Mocha-Dare-7b-ex 在 Open LLM Leaderboard 的评估运行中。
• 数据集结构: 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" 分割始终指向最新结果。
• 额外配置: "results" 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_Azazelle__Mocha-Dare-7b-ex", "harness_winogrande_5", split="train")

最新结果

• 最新结果来源: 2024-03-23T07:33:02.808232
• 结果示例: python { "all": { "acc": 0.6084542013536897, "acc_stderr": 0.0327517442920207, "acc_norm": 0.6113443372287594, "acc_norm_stderr": 0.033399674276373824, "mc1": 0.36107711138310894, "mc1_stderr": 0.016814312844836886, "mc2": 0.5343809110538692, "mc2_stderr": 0.015950754983245087 }, "harness|arc:challenge|25": { "acc": 0.5699658703071673, "acc_stderr": 0.014467631559137986, "acc_norm": 0.6126279863481229, "acc_norm_stderr": 0.014235872487909869 }, "harness|hellaswag|10": { "acc": 0.6310495917147978, "acc_stderr": 0.004815343349305214, "acc_norm": 0.8159729137621987, "acc_norm_stderr": 0.003867143274914471 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.29, "acc_stderr": 0.045604802157206845, "acc_norm": 0.29, "acc_norm_stderr": 0.045604802157206845 }, ... }

配置信息

• 配置名称: harness_arc_challenge_25

  • 数据文件:
    • 分割: 2024_03_23T07_33_02.808232
      • 路径: **/details_harness|arc:challenge|25_2024-03-23T07-33-02.808232.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2024-03-23T07-33-02.808232.parquet

• 配置名称: harness_gsm8k_5

  • 数据文件:
    • 分割: 2024_03_23T07_33_02.808232
      • 路径: **/details_harness|gsm8k|5_2024-03-23T07-33-02.808232.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2024-03-23T07-33-02.808232.parquet

• 配置名称: harness_hellaswag_10

  • 数据文件:
    • 分割: 2024_03_23T07_33_02.808232
      • 路径: **/details_harness|hellaswag|10_2024-03-23T07-33-02.808232.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2024-03-23T07-33-02.808232.parquet

• 配置名称: harness_hendrycksTest_5

  • 数据文件:
    • 分割: 2024_03_23T07_33_02.808232
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2024-03-23T07-33-02.808232.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2024-03-23T07