five

open-llm-leaderboard-old/details_ogimgio__gpt-neo-125m-neurallinguisticpioneers

收藏
Hugging Face2023-10-25 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_ogimgio__gpt-neo-125m-neurallinguisticpioneers
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 ogimgio/gpt-neo-125m-neurallinguisticpioneers 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。它由 2 次运行生成,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 分割始终指向最新结果。一个名为 results 的额外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 中的 datasets 库加载运行细节的示例。

该数据集是在模型 ogimgio/gpt-neo-125m-neurallinguisticpioneers 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。它由 2 次运行生成,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 分割始终指向最新结果。一个名为 results 的额外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 中的 datasets 库加载运行细节的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

该数据集是在对模型 ogimgio/gpt-neo-125m-neurallinguisticpioneers 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集组成

数据集包含64个配置,每个配置对应一个评估任务。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。

加载数据示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_ogimgio__gpt-neo-125m-neurallinguisticpioneers", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-25T03:45:38.309218 运行的最新结果

python { "all": { "em": 0.0024119127516778523, "em_stderr": 0.0005023380498893281, "f1": 0.041171350671141034, "f1_stderr": 0.0012117312202057759, "acc": 0.2564958864222914, "acc_stderr": 0.007403214467064075 }, "harness|drop|3": { "em": 0.0024119127516778523, "em_stderr": 0.0005023380498893281, "f1": 0.041171350671141034, "f1_stderr": 0.0012117312202057759 }, "harness|gsm8k|5": { "acc": 0.000758150113722517, "acc_stderr": 0.0007581501137225334 }, "harness|winogrande|5": { "acc": 0.5122336227308603, "acc_stderr": 0.014048278820405616 } }

配置详情

配置列表

  • harness_arc_challenge_25

    • 分割: 2023_07_19T13_56_53.861726
    • 路径: **/details_harness|arc:challenge|25_2023-07-19T13:56:53.861726.parquet
    • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2023-07-19T13:56:53.861726.parquet
  • harness_drop_3

    • 分割: 2023_10_25T03_45_38.309218
    • 路径: **/details_harness|drop|3_2023-10-25T03-45-38.309218.parquet
    • 分割: latest
    • 路径: **/details_harness|drop|3_2023-10-25T03-45-38.309218.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_25T03_45_38.309218
    • 路径: **/details_harness|gsm8k|5_2023-10-25T03-45-38.309218.parquet
    • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2023-10-25T03-45-38.309218.parquet
  • harness_hellaswag_10

    • 分割: 2023_07_19T13_56_53.861726
    • 路径: **/details_harness|hellaswag|10_2023-07-19T13:56:53.861726.parquet
    • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2023-07-19T13:56:53.861726.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_07_19T13_56_53.861726
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-management|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-07-19T13:56:53.861726.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T13:56:53.861726.parquet
      • `**/details_harness|hendrycksTest-professional_law|5_2023-07-19T13:56:53.8
二维码
社区交流群
二维码
科研交流群
商业服务