five

open-llm-leaderboard-old/details_abhishek__autotrain-8kfjk-b3gva

收藏
Hugging Face2024-03-21 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_abhishek__autotrain-8kfjk-b3gva
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abhishek/autotrain-8kfjk-b3gva的评估运行期间自动创建的,用于Open LLM Leaderboard的评估。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,并用于计算和显示Open LLM Leaderboard上的聚合指标。

该数据集是在模型abhishek/autotrain-8kfjk-b3gva的评估运行期间自动创建的,用于Open LLM Leaderboard的评估。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,并用于计算和显示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

该数据集是在模型abhishek/autotrain-8kfjk-b3gva的评估运行期间自动创建的,用于Open LLM Leaderboard

数据集结构

  • 配置数量: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割存在,分割名称使用运行的时间戳。
  • 训练分割:"train"分割始终指向最新的结果。
  • 结果配置:一个额外的配置"results"存储所有运行的聚合结果,用于计算和显示聚合指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_abhishek__autotrain-8kfjk-b3gva", "harness_winogrande_5", split="train")

最新结果

这些是最新的结果,来自2024-03-21T21:50:17.156229的运行: python { "all": { "acc": 0.5073039070198733, "acc_stderr": 0.03416399026200219, "acc_norm": 0.514093576074635, "acc_norm_stderr": 0.034903503602238484, "mc1": 0.28518971848225216, "mc1_stderr": 0.015805827874454892, "mc2": 0.4349391981377495, "mc2_stderr": 0.01651083406653447 }, "harness|arc:challenge|25": { "acc": 0.4616040955631399, "acc_stderr": 0.014568245550296366, "acc_norm": 0.5, "acc_norm_stderr": 0.014611390804670088 }, "harness|hellaswag|10": { "acc": 0.5385381398127863, "acc_stderr": 0.004974937803907462, "acc_norm": 0.7079267078271261, "acc_norm_stderr": 0.004537865171414024 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.3, "acc_stderr": 0.046056618647183814, "acc_norm": 0.3, "acc_norm_stderr": 0.046056618647183814 }, # 其他任务的结果... }

配置详情

  • harness_arc_challenge_25

    • 分割:2024_03_21T21_48_48.885472, 2024_03_21T21_50_17.156229, latest
    • 路径:**/details_harness|arc:challenge|25_2024-03-21T21-48-48.885472.parquet, **/details_harness|arc:challenge|25_2024-03-21T21-50-17.156229.parquet
  • harness_gsm8k_5

    • 分割:2024_03_21T21_48_48.885472, 2024_03_21T21_50_17.156229, latest
    • 路径:**/details_harness|gsm8k|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|gsm8k|5_2024-03-21T21-50-17.156229.parquet
  • harness_hellaswag_10

    • 分割:2024_03_21T21_48_48.885472, 2024_03_21T21_50_17.156229, latest
    • 路径:**/details_harness|hellaswag|10_2024-03-21T21-48-48.885472.parquet, **/details_harness|hellaswag|10_2024-03-21T21-50-17.156229.parquet
  • harness_hendrycksTest_5

    • 分割:2024_03_21T21_48_48.885472, 2024_03_21T21_50_17.156229, latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-anatomy|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-astronomy|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-business_ethics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-college_biology|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-college_chemistry|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-college_computer_science|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-college_mathematics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-college_medicine|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-college_physics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-computer_security|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-econometrics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-formal_logic|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-global_facts|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_biology|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_geography|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_physics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-human_aging|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-human_sexuality|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-international_law|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-jurisprudence|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-machine_learning|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-management|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-marketing|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-medical_genetics|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-miscellaneous|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-moral_disputes|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-nutrition|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-philosophy|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-prehistory|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-professional_accounting|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-professional_law|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-professional_medicine|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-professional_psychology|5_2024-03-21T21-48-48.885472.parquet, **/details_harness|hendrycksTest-public_relations|5_2024-03-21T21-48-48.885472.parquet, `**/details_harness|hendrycksTest-security_studies|5_2024-03-2
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