five

open-llm-leaderboard-old/details_BryanSwk__LaserPipe-7B-SLERP

收藏
Hugging Face2024-02-09 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_BryanSwk__LaserPipe-7B-SLERP
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BryanSwk/LaserPipe-7B-SLERP进行评估时自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。数据集由2次运行生成,每次运行在每个配置中作为一个特定的分割,使用运行的时间戳命名。train分割始终指向最新的结果。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Open LLM Leaderboard上的聚合指标。

该数据集是在Open LLM Leaderboard上对模型BryanSwk/LaserPipe-7B-SLERP进行评估时自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。数据集由2次运行生成,每次运行在每个配置中作为一个特定的分割,使用运行的时间戳命名。train分割始终指向最新的结果。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

数据集组成

  • 该数据集包含63个配置,每个配置对应一个评估任务。
  • 数据集由2次运行结果组成,每个运行结果作为一个特定的分割(split)存储,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_BryanSwk__LaserPipe-7B-SLERP", "harness_winogrande_5", split="train")

最新结果

  • 以下是最新结果(来自2024-02-09T17:51:06.953880运行)的摘要:

python { "all": { "acc": 0.6543004174004248, "acc_stderr": 0.03203995731196805, "acc_norm": 0.6535960703657506, "acc_norm_stderr": 0.0327123018686762, "mc1": 0.4847001223990208, "mc1_stderr": 0.017495304473187902, "mc2": 0.6537514040125285, "mc2_stderr": 0.014989575626855774 }, "harness|arc:challenge|25": { "acc": 0.6834470989761092, "acc_stderr": 0.013592431519068079, "acc_norm": 0.7107508532423208, "acc_norm_stderr": 0.013250012579393441 }, "harness|hellaswag|10": { "acc": 0.6950806612228639, "acc_stderr": 0.004594323838650357, "acc_norm": 0.8789085839474209, "acc_norm_stderr": 0.0032556675321152896 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.32, "acc_stderr": 0.04688261722621504, "acc_norm": 0.32, "acc_norm_stderr": 0.04688261722621504 }, # 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数据文件路径。以下是部分配置示例:

配置:harness_arc_challenge_25

  • 数据文件:
    • 分割:2024_02_09T17_43_37.094043
      • 路径:**/details_harness|arc:challenge|25_2024-02-09T17-43-37.094043.parquet
    • 分割:2024_02_09T17_51_06.953880
      • 路径:**/details_harness|arc:challenge|25_2024-02-09T17-51-06.953880.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2024-02-09T17-51-06.953880.parquet

配置:harness_gsm8k_5

  • 数据文件:
    • 分割:2024_02_09T17_43_37.094043
      • 路径:**/details_harness|gsm8k|5_2024-02-09T17-43-37.094043.parquet
    • 分割:2024_02_09T17_51_06.953880
      • 路径:**/details_harness|gsm8k|5_2024-02-09T17-51-06.953880.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2024-02-09T17-51-06.953880.parquet

配置:harness_hellaswag_10

  • 数据文件:
    • 分割:2024_02_09T17_43_37.094043
      • 路径:**/details_harness|hellaswag|10_2024-02-09T17-43-37.094043.parquet
    • 分割:2024_02_09T17_51_06.953880
      • 路径:**/details_harness|hellaswag|10_2024-02-09T17-51-06.953880.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2024-02-09T17-51-06.953880.parquet

配置:harness_hendrycksTest_5

  • 数据文件:
    • 分割:2024_02_09T17_43_37.094043
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-international_law|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-management|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-marketing|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2024-02-09T17-43-37.094043.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-09T17-43-37.094043.parquet
        • `**/details_harness|hendrycksTest-nutrition|5_2024-02-09T17-43-3
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