five

open-llm-leaderboard/details_PygmalionAI__mythalion-13b

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_PygmalionAI__mythalion-13b
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PygmalionAI/mythalion-13b进行评估时自动创建的。数据集由64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分割。train分割始终指向最新的结果。此外,还有一个results配置,用于存储聚合结果,这些结果用于在Open LLM Leaderboard上计算和显示指标。README还提供了如何使用`datasets`库加载数据集的示例,并提供了特定运行的最新结果。

该数据集是在Open LLM Leaderboard上对模型PygmalionAI/mythalion-13b进行评估时自动创建的。数据集由64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分割。train分割始终指向最新的结果。此外,还有一个results配置,用于存储聚合结果,这些结果用于在Open LLM Leaderboard上计算和显示指标。README还提供了如何使用`datasets`库加载数据集的示例,并提供了特定运行的最新结果。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集名称

  • Evaluation run of PygmalionAI/mythalion-13b

数据集描述

数据集结构

  • 数据集由64个配置组成,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。
  • 一个额外的配置 "results" 存储所有运行结果的聚合,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_PygmalionAI__mythalion-13b", "harness_winogrande_5", split="train")

最新结果

  • 这些是最新的结果,来自2023-10-26T08:48:40.818758的运行: python { "all": { "em": 0.005243288590604027, "em_stderr": 0.0007396052260778182, "f1": 0.07011430369127479, "f1_stderr": 0.0015312669887699872, "acc": 0.453473099433751, "acc_stderr": 0.010546777696172384 }, "harness|drop|3": { "em": 0.005243288590604027, "em_stderr": 0.0007396052260778182, "f1": 0.07011430369127479, "f1_stderr": 0.0015312669887699872 }, "harness|gsm8k|5": { "acc": 0.1326762699014405, "acc_stderr": 0.009343929131442217 }, "harness|winogrande|5": { "acc": 0.7742699289660616, "acc_stderr": 0.011749626260902552 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|arc:challenge|25_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-09-13T15-43-56.959580.parquet
  • harness_drop_3

    • 分割:2023_10_26T08_48_40.818758
      • 路径:**/details_harness|drop|3_2023-10-26T08-48-40.818758.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-26T08-48-40.818758.parquet
  • harness_gsm8k_5

    • 分割:2023_10_26T08_48_40.818758
      • 路径:**/details_harness|gsm8k|5_2023-10-26T08-48-40.818758.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-26T08-48-40.818758.parquet
  • harness_hellaswag_10

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hellaswag|10_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-management|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_psychology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-public_relations|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-security_studies|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-sociology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-us_foreign_policy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-virology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-world_religions|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-management|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-professional_psychology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-public_relations|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-security_studies|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-sociology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-us_foreign_policy|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-virology|5_2023-09-13T15-43-56.959580.parquet
        • **/details_harness|hendrycksTest-world_religions|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-13T15-43-56.959580.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割:2023_09_13T15_43_56.959580
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-13T15-43-56.959580.parquet
    • 分割:latest
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-13T15-43-56.959580.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