five

open-llm-leaderboard-old/details_perlthoughts__Chupacabra-8x7B-MoE

收藏
Hugging Face2023-12-16 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_perlthoughts__Chupacabra-8x7B-MoE
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型perlthoughts/Chupacabra-8x7B-MoE时自动创建的,主要用于在Open LLM Leaderboard上展示评估结果。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行的结果作为特定配置中的一个分割,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,用于计算和展示Open LLM Leaderboard上的聚合指标。

该数据集是在评估模型perlthoughts/Chupacabra-8x7B-MoE时自动创建的,主要用于在Open LLM Leaderboard上展示评估结果。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行的结果作为特定配置中的一个分割,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,用于计算和展示Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard-old
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 perlthoughts/Chupacabra-8x7B-MoEOpen LLM Leaderboard 上的运行过程中自动创建的。

数据集组成

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。

数据集加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_perlthoughts__Chupacabra-8x7B-MoE", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果 的摘要: python { "all": { "acc": 0.6415463145761969, "acc_stderr": 0.03233222952484684, "acc_norm": 0.6432118874966731, "acc_norm_stderr": 0.03298409149127022, "mc1": 0.47613219094247244, "mc1_stderr": 0.017483547156961578, "mc2": 0.6350369384683723, "mc2_stderr": 0.01508168993616602 }, "harness|arc:challenge|25": { "acc": 0.6561433447098977, "acc_stderr": 0.01388064457015621, "acc_norm": 0.6877133105802048, "acc_norm_stderr": 0.013542598541688067 }, "harness|hellaswag|10": { "acc": 0.6757618004381597, "acc_stderr": 0.004671328673217797, "acc_norm": 0.8610834495120494, "acc_norm_stderr": 0.003451525868724678 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.29, "acc_stderr": 0.04560480215720684, "acc_norm": 0.29, "acc_norm_stderr": 0.04560480215720684 }, # 其他任务的结果... }

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2023_12_16T21_20_16.522598
      • path: **/details_harness|arc:challenge|25_2023-12-16T21-20-16.522598.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-12-16T21-20-16.522598.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_12_16T21_20_16.522598
      • path: **/details_harness|gsm8k|5_2023-12-16T21-20-16.522598.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-12-16T21-20-16.522598.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_12_16T21_20_16.522598
      • path: **/details_harness|hellaswag|10_2023-12-16T21-20-16.522598.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-12-16T21-20-16.522598.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_12_16T21_20_16.522598
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-12-16T21-20-16.522598.parquet

        其他路径...

    • split: latest
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-12-16T21-20-16.522598.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-12-16T21-20-16.522598.parquet

        其他路径...

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