five

open-llm-leaderboard-old/details_ChaoticNeutrals__Kool-Aid_7B

收藏
Hugging Face2024-03-15 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_ChaoticNeutrals__Kool-Aid_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Open LLM Leaderboard上对模型ChaoticNeutrals/Kool-Aid_7B进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集由2次运行创建,每次运行在每个配置中表示为特定的分割,train分割始终指向最新结果。此外,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`datasets`库中的`load_dataset`函数加载运行详情的示例。README中还包含了特定运行的最新结果。

该数据集是在Open LLM Leaderboard上对模型ChaoticNeutrals/Kool-Aid_7B进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集由2次运行创建,每次运行在每个配置中表示为特定的分割,train分割始终指向最新结果。此外,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`datasets`库中的`load_dataset`函数加载运行详情的示例。README中还包含了特定运行的最新结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集信息

  • 数据集名称: Evaluation run of ChaoticNeutrals/Kool-Aid_7B
  • 数据集来源: 自动创建于模型 ChaoticNeutrals/Kool-Aid_7BOpen LLM Leaderboard 的评估运行期间。
  • 数据集组成: 包含63个配置,每个配置对应一个评估任务。
  • 数据集创建: 从2次运行中创建,每次运行作为一个特定的分割,分割名称使用运行的时间戳。"train" 分割始终指向最新结果。
  • 额外配置: "results" 配置存储所有运行的聚合结果,用于计算和显示聚合指标在 Open LLM Leaderboard

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_ChaoticNeutrals__Kool-Aid_7B", "harness_winogrande_5", split="train")

最新结果

  • 最新结果日期: 2024-03-15T06:41:25.894890
  • 结果详情: 包含多个任务的准确率(acc)、标准化准确率(acc_norm)、标准误差(stderr)等指标。

配置详情

  • 配置名称: harness_arc_challenge_25

    • 数据文件:
      • 分割: 2024_03_15T06_36_10.417380
        • 路径: **/details_harness|arc:challenge|25_2024-03-15T06-36-10.417380.parquet
      • 分割: 2024_03_15T06_41_25.894890
        • 路径: **/details_harness|arc:challenge|25_2024-03-15T06-41-25.894890.parquet
      • 分割: latest
        • 路径: **/details_harness|arc:challenge|25_2024-03-15T06-41-25.894890.parquet
  • 配置名称: harness_gsm8k_5

    • 数据文件:
      • 分割: 2024_03_15T06_36_10.417380
        • 路径: **/details_harness|gsm8k|5_2024-03-15T06-36-10.417380.parquet
      • 分割: 2024_03_15T06_41_25.894890
        • 路径: **/details_harness|gsm8k|5_2024-03-15T06-41-25.894890.parquet
      • 分割: latest
        • 路径: **/details_harness|gsm8k|5_2024-03-15T06-41-25.894890.parquet
  • 配置名称: harness_hellaswag_10

    • 数据文件:
      • 分割: 2024_03_15T06_36_10.417380
        • 路径: **/details_harness|hellaswag|10_2024-03-15T06-36-10.417380.parquet
      • 分割: 2024_03_15T06_41_25.894890
        • 路径: **/details_harness|hellaswag|10_2024-03-15T06-41-25.894890.parquet
      • 分割: latest
        • 路径: **/details_harness|hellaswag|10_2024-03-15T06-41-25.894890.parquet
  • 配置名称: harness_hendrycksTest_5

    • 数据文件:
      • 分割: 2024_03_15T06_36_10.417380
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-management|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-professional_medicine|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-professional_psychology|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-public_relations|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-security_studies|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-sociology|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-us_foreign_policy|5_2024-03-15T06-36-10.417380.parquet
          • **/details_harness|hendrycksTest-virology|5_2024-03-15T06-36-10.4173
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