five

open-llm-leaderboard/details_JosephusCheung__Pwen-7B-Chat-20_30

收藏
Hugging Face2023-10-26 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_JosephusCheung__Pwen-7B-Chat-20_30
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JosephusCheung/Pwen-7B-Chat-20_30在Open LLM Leaderboard上的评估运行期间自动创建的。数据集由64个配置组成,每个配置对应一个被评估的任务。它由2次运行创建,每次运行作为每个配置中的一个特定分割,使用运行的时间戳命名。train分割始终指向最新的结果。一个额外的配置results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还提供了一个使用Python中的datasets库加载运行细节的示例。
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 JosephusCheung/Pwen-7B-Chat-20_30Open LLM Leaderboard 上的自动创建的。数据集包含 64 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

数据集包含以下配置:

  • harness_arc_challenge_25

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|arc:challenge|25_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-10-10T07-01-15.573690.parquet
  • harness_drop_3

    • 分割:2023_10_26T02_42_36.258115
    • 路径:**/details_harness|drop|3_2023-10-26T02-42-36.258115.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-10-26T02-42-36.258115.parquet
  • harness_gsm8k_5

    • 分割:2023_10_26T02_42_36.258115
    • 路径:**/details_harness|gsm8k|5_2023-10-26T02-42-36.258115.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-26T02-42-36.258115.parquet
  • harness_hellaswag_10

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hellaswag|10_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T07-01-15.573690.parquet 等 40 个文件
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T07-01-15.573690.parquet 等 40 个文件
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T07-01-15.573690.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割:2023_10_10T07_01_15.573690
    • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-10T07-01-15.573690.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-10T07-01-15.573690.parquet

最新结果

以下是 2023-10-26T02:42:36.258115 运行 的最新结果:

python { "all": { "em": 0.2954068791946309, "em_stderr": 0.004672175556184236, "f1": 0.3814209312080561, "f1_stderr": 0.004573085663083055, "acc": 0.44525521893903264, "acc_stderr": 0.012103729416391124 }, "harness|drop|3": { "em": 0.2954068791946309, "em_stderr": 0.004672175556184236, "f1": 0.3814209312080561, "f1_stderr": 0.004573085663083055 }, "harness|gsm8k|5": { "acc": 0.20621683093252463, "acc_stderr": 0.0111443640

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