five

open-llm-leaderboard/details_chatty123__mistral_rank32_dpo

收藏
Hugging Face2024-04-15 更新2024-06-12 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_chatty123__mistral_rank32_dpo
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chatty123/mistral_rank32_dpo进行评估时自动生成的。数据集由63个配置组成,每个配置对应一个评估任务。它包含2次运行的数据,每次运行在每个配置中表示为特定的分割,分割名称根据运行的时间戳命名。train分割始终指向最新的结果。此外,还有一个results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Hugging Face datasets库加载数据集的示例。

该数据集是在Open LLM Leaderboard上对模型chatty123/mistral_rank32_dpo进行评估时自动生成的。数据集由63个配置组成,每个配置对应一个评估任务。它包含2次运行的数据,每次运行在每个配置中表示为特定的分割,分割名称根据运行的时间戳命名。train分割始终指向最新的结果。此外,还有一个results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

数据集摘要

该数据集是在评估模型 chatty123/mistral_rank32_dpoOpen LLM Leaderboard 上的自动创建的。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,每个运行可以在每个配置中作为一个特定的分割找到,分割名称使用运行的时间戳。"train" 分割始终指向最新的结果。

最新结果

以下是 最新结果 的摘要:

python { "all": { "acc": 0.6048858578613187, "acc_stderr": 0.033294639924554184, "acc_norm": 0.6097911403143494, "acc_norm_stderr": 0.033969901287993785, "mc1": 0.5263157894736842, "mc1_stderr": 0.017479241161975453, "mc2": 0.6833350117330134, "mc2_stderr": 0.015252782890753571 }, "harness|arc:challenge|25": { "acc": 0.5853242320819113, "acc_stderr": 0.014397070564409174, "acc_norm": 0.6331058020477816, "acc_norm_stderr": 0.014084133118104303 }, "harness|hellaswag|10": { "acc": 0.6654052977494523, "acc_stderr": 0.004708842600177433, "acc_norm": 0.8503286197968533, "acc_norm_stderr": 0.00356019918548657 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.04793724854411021, "acc_norm": 0.35, "acc_norm_stderr": 0.04793724854411021 }, # 其他任务的结果省略... }

配置详情

数据集包含多个配置,每个配置对应不同的评估任务和运行结果。以下是部分配置的示例:

  • config_name: harness_arc_challenge_25

    • 分割: 2024_04_15T18_57_17.206294
      • 路径: **/details_harness|arc:challenge|25_2024-04-15T18-57-17.206294.parquet
    • 分割: 2024_04_15T19_05_28.402145
      • 路径: **/details_harness|arc:challenge|25_2024-04-15T19-05-28.402145.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2024-04-15T19-05-28.402145.parquet
  • config_name: harness_gsm8k_5

    • 分割: 2024_04_15T18_57_17.206294
      • 路径: **/details_harness|gsm8k|5_2024-04-15T18-57-17.206294.parquet
    • 分割: 2024_04_15T19_05_28.402145
      • 路径: **/details_harness|gsm8k|5_2024-04-15T19-05-28.402145.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2024-04-15T19-05-28.402145.parquet
  • config_name: harness_hellaswag_10

    • 分割: 2024_04_15T18_57_17.206294
      • 路径: **/details_harness|hellaswag|10_2024-04-15T18-57-17.206294.parquet
    • 分割: 2024_04_15T19_05_28.402145
      • 路径: **/details_harness|hellaswag|10_2024-04-15T19-05-28.402145.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2024-04-15T19-05-28.402145.parquet
  • config_name: harness_hendrycksTest_5

    • 分割: 2024_04_15T18_57_17.206294
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-04-15T18-57-17.206294.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-15T18-57-17.206294.parquet
    • 分割: 2024_04_15T19_05_28.402145
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-15T19-05-28.402145.parquet
    • 分割: latest
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-04-15T19-05-28.402145.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-15T19-05-28.402145.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