five

open-llm-leaderboard-old/details_Aryanne__YarnLake-Swap-7B

收藏
Hugging Face2024-02-04 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Aryanne__YarnLake-Swap-7B
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型Aryanne/YarnLake-Swap-7B评估运行期间自动创建的,用于Open LLM排行榜。数据集包含63个配置,每个配置对应一个评估任务。数据集来自1次运行,每次运行作为一个特定分割,以运行时间戳命名。train分割始终指向最新结果。额外配置results存储了运行的所有聚合结果,用于计算和显示排行榜上的聚合指标。数据集针对不同任务有特定配置,每个配置包含针对不同分割(包括最新结果)的数据文件。

该数据集是在模型Aryanne/YarnLake-Swap-7B评估运行期间自动创建的,用于Open LLM排行榜。数据集包含63个配置,每个配置对应一个评估任务。数据集来自1次运行,每次运行作为一个特定分割,以运行时间戳命名。train分割始终指向最新结果。额外配置results存储了运行的所有聚合结果,用于计算和显示排行榜上的聚合指标。数据集针对不同任务有特定配置,每个配置包含针对不同分割(包括最新结果)的数据文件。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 Aryanne/YarnLake-Swap-7B 进行评估运行时自动创建的,用于 Open LLM Leaderboard

数据集结构

数据集包含 63 个配置,每个配置对应一个评估任务。数据集从 1 次运行中创建,每个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。

数据集加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Aryanne__YarnLake-Swap-7B", "harness_winogrande_5", split="train")

最新结果

以下是 2024-02-04T06:28:34.013587 运行的最新结果

python { "all": { "acc": 0.6487764218521861, "acc_stderr": 0.03199354917196398, "acc_norm": 0.6513693263333427, "acc_norm_stderr": 0.03263625404197957, "mc1": 0.3353733170134639, "mc1_stderr": 0.01652753403966899, "mc2": 0.490726463865428, "mc2_stderr": 0.014923584147103909 }, "harness|arc:challenge|25": { "acc": 0.6194539249146758, "acc_stderr": 0.014188277712349815, "acc_norm": 0.6527303754266212, "acc_norm_stderr": 0.01391303452962045 }, "harness|hellaswag|10": { "acc": 0.663114917347142, "acc_stderr": 0.004716792874433208, "acc_norm": 0.8517227643895638, "acc_norm_stderr": 0.0035464830155691185 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.33, "acc_stderr": 0.04725815626252606, "acc_norm": 0.33, "acc_norm_stderr": 0.04725815626252606 }, # 其他任务的结果... }

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2024_02_04T06_28_34.013587
      • path: **/details_harness|arc:challenge|25_2024-02-04T06-28-34.013587.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2024-02-04T06-28-34.013587.parquet
  • config_name: harness_gsm8k_5

    • split: 2024_02_04T06_28_34.013587
      • path: **/details_harness|gsm8k|5_2024-02-04T06-28-34.013587.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2024-02-04T06-28-34.013587.parquet
  • config_name: harness_hellaswag_10

    • split: 2024_02_04T06_28_34.013587
      • path: **/details_harness|hellaswag|10_2024-02-04T06-28-34.013587.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2024-02-04T06-28-34.013587.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2024_02_04T06_28_34.013587
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-international_law|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-management|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-marketing|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2024-02-04T06-28-34.013587.parquet
        • **/details_harness|hendrycksTest-professional_psychology|5_20
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