five

open-llm-leaderboard-old/details_yam-peleg__Experiment31-7B

收藏
Hugging Face2024-03-10 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_yam-peleg__Experiment31-7B
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 yam-peleg/Experiment31-7B 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 63 个配置组成,每个配置对应一个评估任务。数据集是从 2 次运行中生成的,每次运行在每个配置中表示为特定的拆分,train 拆分始终指向最新结果。一个名为 results 的额外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。README 还提供了如何使用 `datasets` 库中的 `load_dataset` 函数加载数据集的示例。

该数据集是在模型 yam-peleg/Experiment31-7B 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 63 个配置组成,每个配置对应一个评估任务。数据集是从 2 次运行中生成的,每次运行在每个配置中表示为特定的拆分,train 拆分始终指向最新结果。一个名为 results 的额外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。README 还提供了如何使用 `datasets` 库中的 `load_dataset` 函数加载数据集的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

该数据集是在评估模型 yam-peleg/Experiment31-7B 的过程中自动创建的,用于 Open LLM Leaderboard。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 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。

数据集结构

配置

  • config_name: harness_arc_challenge_25

    • data_files:
      • split: 2024_03_07T12_02_53.998905
        • path: **/details_harness|arc:challenge|25_2024-03-07T12-02-53.998905.parquet
      • split: 2024_03_10T00_56_13.235163
        • path: **/details_harness|arc:challenge|25_2024-03-10T00-56-13.235163.parquet
      • split: latest
        • path: **/details_harness|arc:challenge|25_2024-03-10T00-56-13.235163.parquet
  • config_name: harness_gsm8k_5

    • data_files:
      • split: 2024_03_07T12_02_53.998905
        • path: **/details_harness|gsm8k|5_2024-03-07T12-02-53.998905.parquet
      • split: 2024_03_10T00_56_13.235163
        • path: **/details_harness|gsm8k|5_2024-03-10T00-56-13.235163.parquet
      • split: latest
        • path: **/details_harness|gsm8k|5_2024-03-10T00-56-13.235163.parquet
  • config_name: harness_hellaswag_10

    • data_files:
      • split: 2024_03_07T12_02_53.998905
        • path: **/details_harness|hellaswag|10_2024-03-07T12-02-53.998905.parquet
      • split: 2024_03_10T00_56_13.235163
        • path: **/details_harness|hellaswag|10_2024-03-10T00-56-13.235163.parquet
      • split: latest
        • path: **/details_harness|hellaswag|10_2024-03-10T00-56-13.235163.parquet
  • config_name: harness_hendrycksTest_5

    • data_files:
      • split: 2024_03_07T12_02_53.998905
        • path:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-03-07T12-02-53.998905.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-07T12-02-53.998905.parquet
      • split: 2024_03_10T00_56_13.235163
        • path:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-10T00-56-13.235163.parquet
      • split: latest
        • path:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-03-10T00-56-13.235163.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-10T00-56-13.235163.parquet

最新结果

以下是 2024-03-10T00:56:13.235163 运行 的最新结果:

python { "all": { "acc": 0.6501244499294567, "acc_stderr": 0.03210723010886301, "acc_norm": 0.6493350826485453, "acc_norm_stderr": 0.032781212255641866, "mc1": 0.6328029375764994, "mc1_stderr": 0.016874805001453184, "mc2": 0.7831336319427918, "mc2_stderr": 0.013608643125458352 }, "harness|arc:challenge|25": { "acc": 0.7158703071672355, "acc_stderr": 0.013179442447653884, "acc_norm": 0.735494880546075, "acc_norm_stderr": 0.012889272949313368 }, "harness|hellaswag|10": { "acc": 0.7175861382194781, "acc_stderr": 0.004492535748097628, "acc_norm": 0.8918542123083051, "acc_norm_stderr": 0.0030992974183235455 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.32, "acc_stderr": 0.046882617226215034, "acc_norm": 0.32, "acc_norm_stderr": 0.046882617226215034 }, "harness|hendrycksTest-anatomy|5": { "acc": 0.6370370370370371, "acc_stderr": 0.04153948404742398, "acc_norm": 0.63

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