five

open-llm-leaderboard/details_icefog72__IceTeaRP-7b

收藏
Hugging Face2024-03-28 更新2024-06-11 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_icefog72__IceTeaRP-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icefog72/IceTeaRP-7b时自动创建的,主要用于在Open LLM Leaderboard上展示评估结果。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,每次运行的结果存储为特定的分割,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,用于在Open LLM Leaderboard上计算和展示聚合指标。

该数据集是在评估模型icefog72/IceTeaRP-7b时自动创建的,主要用于在Open LLM Leaderboard上展示评估结果。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,每次运行的结果存储为特定的分割,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,用于在Open LLM Leaderboard上计算和展示聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 icefog72/IceTeaRP-7bOpen LLM Leaderboard 上的自动创建的。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

  • 配置数量:63
  • 创建来源:1 次运行
  • 数据分割:每个配置包含特定分割,分割名称使用运行的时间戳。"train" 分割始终指向最新结果。
  • 额外配置:"results" 存储所有运行结果的聚合,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_icefog72__IceTeaRP-7b", "harness_winogrande_5", split="train")

最新结果

最新结果来自 2024-03-28T15:07:11.713345 运行,包含多个任务的评估结果。以下是部分结果示例:

python { "all": { "acc": 0.6420523435791596, "acc_stderr": 0.03245624935569837, "acc_norm": 0.6439245877855285, "acc_norm_stderr": 0.03310750757095215, "mc1": 0.4467564259485924, "mc1_stderr": 0.017403977522557144, "mc2": 0.6243602183451233, "mc2_stderr": 0.015478986567917703 }, "harness|arc:challenge|25": { "acc": 0.636518771331058, "acc_stderr": 0.014056207319068287, "acc_norm": 0.6697952218430034, "acc_norm_stderr": 0.013743085603760422 }, "harness|hellaswag|10": { "acc": 0.6777534355706034, "acc_stderr": 0.004663817291468729, "acc_norm": 0.8612826130252937, "acc_norm_stderr": 0.00344944961865055 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.33, "acc_stderr": 0.04725815626252606, "acc_norm": 0.33, "acc_norm_stderr": 0.04725815626252606 }, # 更多任务结果... }

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:

  • 配置名称:harness_arc_challenge_25

    • 数据文件
      • 分割:2024_03_28T15_07_11.713345
        • 路径:**/details_harness|arc:challenge|25_2024-03-28T15-07-11.713345.parquet
      • 分割:latest
        • 路径:**/details_harness|arc:challenge|25_2024-03-28T15-07-11.713345.parquet
  • 配置名称:harness_gsm8k_5

    • 数据文件
      • 分割:2024_03_28T15_07_11.713345
        • 路径:**/details_harness|gsm8k|5_2024-03-28T15-07-11.713345.parquet
      • 分割:latest
        • 路径:**/details_harness|gsm8k|5_2024-03-28T15-07-11.713345.parquet
  • 配置名称:harness_hellaswag_10

    • 数据文件
      • 分割:2024_03_28T15_07_11.713345
        • 路径:**/details_harness|hellaswag|10_2024-03-28T15-07-11.713345.parquet
      • 分割:latest
        • 路径:**/details_harness|hellaswag|10_2024-03-28T15-07-11.713345.parquet
  • 配置名称:harness_hendrycksTest_5

    • 数据文件
      • 分割:2024_03_28T15_07_11.713345
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-management|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-03-28T15-07-11.713345.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-03-28T15-07-11.713345.parquet
          • `**/details_harness|hendrycksTest-professional_law|5_2024-03-28T15-07-11.713345.parquet
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