five

open-llm-leaderboard-old/details_macadeliccc__polyglot-math-4x7b

收藏
Hugging Face2024-01-14 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_macadeliccc__polyglot-math-4x7b
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macadeliccc/polyglot-math-4x7b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,每次运行的结果作为特定配置中的一个分割,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,用于计算和展示在Open LLM Leaderboard上的聚合指标。

该数据集是在评估模型macadeliccc/polyglot-math-4x7b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,每次运行的结果作为特定配置中的一个分割,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,用于计算和展示在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

该数据集是在评估模型 macadeliccc/polyglot-math-4x7bOpen 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
  • 数据来源: 1 次运行
  • 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" 分割指向最新结果。
  • 额外配置: "results" 存储所有聚合结果,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_macadeliccc__polyglot-math-4x7b", "harness_winogrande_5", split="train")

最新结果

以下是 2024-01-14T01:25:55.830403 的最新结果:

python { "all": { "acc": 0.6367747877161951, "acc_stderr": 0.03232816338890694, "acc_norm": 0.6393383626953215, "acc_norm_stderr": 0.03297276004070419, "mc1": 0.3635250917992656, "mc1_stderr": 0.016838862883965827, "mc2": 0.5378477391082209, "mc2_stderr": 0.015247687104643274 }, "harness|arc:challenge|25": { "acc": 0.6023890784982935, "acc_stderr": 0.014301752223279542, "acc_norm": 0.6373720136518771, "acc_norm_stderr": 0.014049106564955009 }, "harness|hellaswag|10": { "acc": 0.6549492133041227, "acc_stderr": 0.004744132825391518, "acc_norm": 0.8485361481776539, "acc_norm_stderr": 0.0035776774950640874 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.29, "acc_stderr": 0.045604802157206845, "acc_norm": 0.29, "acc_norm_stderr": 0.045604802157206845 }, # 其他任务的结果... }

配置详情

  • 配置名称: harness_arc_challenge_25

    • 数据文件:
      • 分割: 2024_01_14T01_25_55.830403
        • 路径: **/details_harness|arc:challenge|25_2024-01-14T01-25-55.830403.parquet
      • 分割: latest
        • 路径: **/details_harness|arc:challenge|25_2024-01-14T01-25-55.830403.parquet
  • 配置名称: harness_gsm8k_5

    • 数据文件:
      • 分割: 2024_01_14T01_25_55.830403
        • 路径: **/details_harness|gsm8k|5_2024-01-14T01-25-55.830403.parquet
      • 分割: latest
        • 路径: **/details_harness|gsm8k|5_2024-01-14T01-25-55.830403.parquet
  • 配置名称: harness_hellaswag_10

    • 数据文件:
      • 分割: 2024_01_14T01_25_55.830403
        • 路径: **/details_harness|hellaswag|10_2024-01-14T01-25-55.830403.parquet
      • 分割: latest
        • 路径: **/details_harness|hellaswag|10_2024-01-14T01-25-55.830403.parquet
  • 配置名称: harness_hendrycksTest_5

    • 数据文件:
      • 分割: 2024_01_14T01_25_55.830403
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-management|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-01-14T01-25-55.830403.parquet
          • **/details_harness|hend
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