five

open-llm-leaderboard-old/details_Kabster__Bio-Mistralv2-Squared

收藏
Hugging Face2024-03-10 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Kabster__Bio-Mistralv2-Squared
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Kabster/Bio-Mistralv2-Squared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,每次运行作为每个配置中的特定拆分,拆分名称使用运行的时间戳。train拆分始终指向最新结果。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Python代码加载运行细节的示例,并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Kabster/Bio-Mistralv2-Squared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,每次运行作为每个配置中的特定拆分,拆分名称使用运行的时间戳。train拆分始终指向最新结果。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Python代码加载运行细节的示例,并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

数据集简介

该数据集是在模型 Kabster/Bio-Mistralv2-SquaredOpen LLM Leaderboard 上的评估运行期间自动创建的。数据集包含 63 个配置,每个配置对应一个评估任务。

数据集结构

数据集从 2 次运行中创建,每次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。

额外配置

一个额外的配置 "results" 存储了所有运行的聚合结果,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_Kabster__Bio-Mistralv2-Squared", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果 的摘要: python { "all": { "acc": 0.5947088232667346, "acc_stderr": 0.033001873861023094, "acc_norm": 0.605373945602307, "acc_norm_stderr": 0.033881891438851675, "mc1": 0.42962056303549573, "mc1_stderr": 0.0173292345804091, "mc2": 0.6098919620417469, "mc2_stderr": 0.015392589865179624 }, "harness|arc:challenge|25": { "acc": 0.5938566552901023, "acc_stderr": 0.014351656690097862, "acc_norm": 0.6296928327645052, "acc_norm_stderr": 0.01411129875167495 }, "harness|hellaswag|10": { "acc": 0.657239593706433, "acc_stderr": 0.004736621698861176, "acc_norm": 0.8401712806213901, "acc_norm_stderr": 0.0036569821653861666 }, "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

数据集包含多个配置,每个配置对应不同的任务和分割。以下是部分配置的示例:

  • config_name: harness_arc_challenge_25

    • 分割: 2024_03_09T23_11_18.304357
      • 路径: **/details_harness|arc:challenge|25_2024-03-09T23-11-18.304357.parquet
    • 分割: 2024_03_10T05_14_57.667370
      • 路径: **/details_harness|arc:challenge|25_2024-03-10T05-14-57.667370.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2024-03-10T05-14-57.667370.parquet
  • config_name: harness_gsm8k_5

    • 分割: 2024_03_09T23_11_18.304357
      • 路径: **/details_harness|gsm8k|5_2024-03-09T23-11-18.304357.parquet
    • 分割: 2024_03_10T05_14_57.667370
      • 路径: **/details_harness|gsm8k|5_2024-03-10T05-14-57.667370.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2024-03-10T05-14-57.667370.parquet
  • config_name: harness_hellaswag_10

    • 分割: 2024_03_09T23_11_18.304357
      • 路径: **/details_harness|hellaswag|10_2024-03-09T23-11-18.304357.parquet
    • 分割: 2024_03_10T05_14_57.667370
      • 路径: **/details_harness|hellaswag|10_2024-03-10T05-14-57.667370.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2024-03-10T05-14-57.667370.parquet
  • config_name: harness_hendrycksTest_5

    • 分割: 2024_03_09T23_11_18.304357
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-international_law|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-management|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-marketing|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2024-03-09T23-11-18.304357.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-09T23-11-18.304357.
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