five

open-llm-leaderboard-old/details_manishiitg__open-aditi-hi-v1

收藏
Hugging Face2024-01-05 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_manishiitg__open-aditi-hi-v1
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型manishiitg/open-aditi-hi-v1在Open LLM Leaderboard上的评估运行期间自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。数据集是从1次运行中生成的,每次运行在每个配置中表示为特定的分割,使用运行的时间戳命名。train分割始终指向最新的结果。一个名为results的额外配置存储了运行的所有聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用Python中的datasets库加载该数据集,如提供的代码片段所示。

该数据集是在模型manishiitg/open-aditi-hi-v1在Open LLM Leaderboard上的评估运行期间自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。数据集是从1次运行中生成的,每次运行在每个配置中表示为特定的分割,使用运行的时间戳命名。train分割始终指向最新的结果。一个名为results的额外配置存储了运行的所有聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用Python中的datasets库加载该数据集,如提供的代码片段所示。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集信息

  • 名称: Evaluation run of manishiitg/open-aditi-hi-v1
  • 来源: 自动创建于模型 manishiitg/open-aditi-hi-v1Open LLM Leaderboard 的评估运行期间。
  • 组成: 包含 63 个配置,每个配置对应一个评估任务。
  • 创建: 从 1 次运行中创建,每次运行的详细结果存储在特定的分割中,分割名称使用运行的时间戳。
  • 最新结果: "train" 分割始终指向最新结果。
  • 聚合结果: 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_manishiitg__open-aditi-hi-v1", "harness_winogrande_5", split="train")

最新结果

  • 来源: 来自运行 2024-01-05T00:54:17.291176 的结果。
  • 详细结果: python { "all": { "acc": 0.583972416790564, "acc_stderr": 0.033496039335568516, "acc_norm": 0.5889834623305892, "acc_norm_stderr": 0.03419061870668813, "mc1": 0.2802937576499388, "mc1_stderr": 0.01572313952460875, "mc2": 0.42336343098693335, "mc2_stderr": 0.014576483672407868 }, "harness|arc:challenge|25": { "acc": 0.5563139931740614, "acc_stderr": 0.01451842182567045, "acc_norm": 0.5878839590443686, "acc_norm_stderr": 0.014383915302225402 }, "harness|hellaswag|10": { "acc": 0.6185022903804023, "acc_stderr": 0.004847615216473459, "acc_norm": 0.8137821151165107, "acc_norm_stderr": 0.003884868131822894 }, ... }

配置信息

  • 配置名称: harness_arc_challenge_25

    • 数据文件:
      • 分割: 2024_01_05T00_54_17.291176
        • 路径: **/details_harness|arc:challenge|25_2024-01-05T00-54-17.291176.parquet
      • 分割: latest
        • 路径: **/details_harness|arc:challenge|25_2024-01-05T00-54-17.291176.parquet
  • 配置名称: harness_gsm8k_5

    • 数据文件:
      • 分割: 2024_01_05T00_54_17.291176
        • 路径: **/details_harness|gsm8k|5_2024-01-05T00-54-17.291176.parquet
      • 分割: latest
        • 路径: **/details_harness|gsm8k|5_2024-01-05T00-54-17.291176.parquet
  • 配置名称: harness_hellaswag_10

    • 数据文件:
      • 分割: 2024_01_05T00_54_17.291176
        • 路径: **/details_harness|hellaswag|10_2024-01-05T00-54-17.291176.parquet
      • 分割: latest
        • 路径: **/details_harness|hellaswag|10_2024-01-05T00-54-17.291176.parquet
  • 配置名称: harness_hendrycksTest_5

    • 数据文件:
      • 分割: 2024_01_05T00_54_17.291176
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-management|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-01-05T00-54-17.291176.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2024-01-05T00-54-17.291176.parquet
          • `**/details_harness|hendrycksTest-professional_medicine|5_2024-01-05T00-
二维码
社区交流群
二维码
科研交流群
商业服务