five

open-llm-leaderboard-old/details_Kukedlc__NeuralMaths-Experiment-7b

收藏
Hugging Face2024-03-29 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Kukedlc__NeuralMaths-Experiment-7b
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型Kukedlc/NeuralMaths-Experiment-7b时自动创建的,主要用于在Open LLM Leaderboard上进行评估。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行的结果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,用于计算和显示Open LLM Leaderboard上的聚合指标。

该数据集是在评估模型Kukedlc/NeuralMaths-Experiment-7b时自动创建的,主要用于在Open LLM Leaderboard上进行评估。数据集包含63个配置,每个配置对应一个评估任务。数据集由1次运行生成,每次运行的结果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,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

数据集摘要

该数据集是在对模型 Kukedlc/NeuralMaths-Experiment-7b 进行评估运行时自动创建的,用于 Open LLM Leaderboard

数据集组成

  • 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拆分,拆分名称使用运行的时间戳。
  • "train" 拆分始终指向最新的结果。
  • 额外的 "results" 配置存储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。

最新结果

以下是 2024-03-29T21:48:56.204016 运行的最新结果

python { "all": { "acc": 0.6556680183910659, "acc_stderr": 0.03185357048069436, "acc_norm": 0.6547192071330743, "acc_norm_stderr": 0.03252307813134486, "mc1": 0.47123623011015914, "mc1_stderr": 0.01747451384852552, "mc2": 0.6382648558896807, "mc2_stderr": 0.015184759114654476 }, "harness|arc:challenge|25": { "acc": 0.6680887372013652, "acc_stderr": 0.013760988200880538, "acc_norm": 0.697098976109215, "acc_norm_stderr": 0.013428241573185349 }, "harness|hellaswag|10": { "acc": 0.6942840071698865, "acc_stderr": 0.004597684609707823, "acc_norm": 0.8748257319259112, "acc_norm_stderr": 0.0033024011069263223 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.31, "acc_stderr": 0.04648231987117316, "acc_norm": 0.31, "acc_norm_stderr": 0.04648231987117316 }, # 其他任务的结果... }

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2024_03_29T21_48_56.204016
      • path: **/details_harness|arc:challenge|25_2024-03-29T21-48-56.204016.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2024-03-29T21-48-56.204016.parquet
  • config_name: harness_gsm8k_5

    • split: 2024_03_29T21_48_56.204016
      • path: **/details_harness|gsm8k|5_2024-03-29T21-48-56.204016.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2024-03-29T21-48-56.204016.parquet
  • config_name: harness_hellaswag_10

    • split: 2024_03_29T21_48_56.204016
      • path: **/details_harness|hellaswag|10_2024-03-29T21-48-56.204016.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2024-03-29T21-48-56.204016.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2024_03_29T21_48_56.204016
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2024-03-29T21-48-56.204016.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2024-03-29T21-48-56.204016.parquet

        其他路径...

}

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