five

open-llm-leaderboard-old/details_CultriX__NeuralMona_MoE-4x7B

收藏
Hugging Face2024-03-27 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_CultriX__NeuralMona_MoE-4x7B
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CultriX/NeuralMona_MoE-4x7B在Open LLM Leaderboard上的评估运行期间自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。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。一个名为results的额外配置存储了运行的所有聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。

该数据集是在模型CultriX/NeuralMona_MoE-4x7B在Open LLM Leaderboard上的评估运行期间自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。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。一个名为results的额外配置存储了运行的所有聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集来源

数据集结构

  • 配置数量:数据集包含63个配置,每个配置对应一个评估任务。
  • 运行次数:数据集从2次运行中创建。每个运行在每个配置中作为一个特定的分片存在,分片名称使用运行的时间戳。
  • 最新结果:"train"分片始终指向最新的结果。
  • 汇总结果:一个额外的配置"results"存储所有运行的汇总结果,用于计算和显示在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_CultriX__NeuralMona_MoE-4x7B", "harness_winogrande_5", split="train")

最新结果

  • 最新结果时间戳:2024-03-27T21:11:00.982952
  • 结果示例: python { "all": { "acc": 0.6491773099838035, "acc_stderr": 0.03216083681682968, "acc_norm": 0.6489140301517058, "acc_norm_stderr": 0.03282766114720701, "mc1": 0.6242350061199511, "mc1_stderr": 0.01695458406021429, "mc2": 0.7727167107815722, "mc2_stderr": 0.013854380605408638 }, "harness|arc:challenge|25": { "acc": 0.71160409556314, "acc_stderr": 0.013238394422428171, "acc_norm": 0.7389078498293515, "acc_norm_stderr": 0.012835523909473833 }, "harness|hellaswag|10": { "acc": 0.7167894841665007, "acc_stderr": 0.004496369742132101, "acc_norm": 0.890161322445728, "acc_norm_stderr": 0.003120495238827556 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.34, "acc_stderr": 0.04760952285695235, "acc_norm": 0.34, "acc_norm_stderr": 0.04760952285695235 },

    其他任务结果省略...

}

配置详情

  • 配置名称:harness_arc_challenge_25

    • 分片
      • 2024_03_27T19_44_07.591584
      • 2024_03_27T21_11_00.982952
      • latest
    • 路径
      • **/details_harness|arc:challenge|25_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|arc:challenge|25_2024-03-27T21-11-00.982952.parquet
  • 配置名称:harness_gsm8k_5

    • 分片
      • 2024_03_27T19_44_07.591584
      • 2024_03_27T21_11_00.982952
      • latest
    • 路径
      • **/details_harness|gsm8k|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|gsm8k|5_2024-03-27T21-11-00.982952.parquet
  • 配置名称:harness_hellaswag_10

    • 分片
      • 2024_03_27T19_44_07.591584
      • 2024_03_27T21_11_00.982952
      • latest
    • 路径
      • **/details_harness|hellaswag|10_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hellaswag|10_2024-03-27T21-11-00.982952.parquet
  • 配置名称:harness_hendrycksTest_5

    • 分片
      • 2024_03_27T19_44_07.591584
      • 2024_03_27T21_11_00.982952
      • latest
    • 路径
      • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-international_law|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-management|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-marketing|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2024-03-27T19-44-07.591584.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2024-03-27T19-44-07.
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