five

open-llm-leaderboard-old/details_uni-tianyan__Uni-TianYan

收藏
Hugging Face2023-09-18 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_uni-tianyan__Uni-TianYan
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uni-tianyan/Uni-TianYan进行评估时自动创建的。数据集由64个配置组成,每个配置对应一个评估任务。数据集由2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,用于在Open LLM Leaderboard上计算和显示聚合指标。

该数据集是在Open LLM Leaderboard上对模型uni-tianyan/Uni-TianYan进行评估时自动创建的。数据集由64个配置组成,每个配置对应一个评估任务。数据集由2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,用于在Open LLM Leaderboard上计算和显示聚合指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

该数据集是在评估模型 uni-tianyan/Uni-TianYanOpen LLM Leaderboard 上的自动创建的。数据集包含 64 个配置,每个配置对应一个评估任务。

数据集结构

  • 配置数量:64 个配置
  • 数据来源:从 2 次运行中创建,每次运行对应一个特定的时间戳
  • 数据格式:每个配置包含多个数据文件,格式为 Parquet

数据集加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_uni-tianyan__Uni-TianYan", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-18T02:40:09.826211 运行的最新结果

python { "all": { "em": 0.24486157718120805, "em_stderr": 0.004403654691385411, "f1": 0.39787751677852523, "f1_stderr": 0.004155160727794137, "acc": 0.5222921265482389, "acc_stderr": 0.01107896164608613 }, "harness|drop|3": { "em": 0.24486157718120805, "em_stderr": 0.004403654691385411, "f1": 0.39787751677852523, "f1_stderr": 0.004155160727794137 }, "harness|gsm8k|5": { "acc": 0.221379833206975, "acc_stderr": 0.011436000004253518 }, "harness|winogrande|5": { "acc": 0.8232044198895028, "acc_stderr": 0.010721923287918744 } }

配置详情

  • harness_arc_challenge_25

    • 数据文件路径:
      • **/details_harness|arc:challenge|25_2023-09-03T12:27:36.436118.parquet
  • harness_drop_3

    • 数据文件路径:
      • **/details_harness|drop|3_2023-09-18T02-40-09.826211.parquet
  • harness_gsm8k_5

    • 数据文件路径:
      • **/details_harness|gsm8k|5_2023-09-18T02-40-09.826211.parquet
  • harness_hellaswag_10

    • 数据文件路径:
      • **/details_harness|hellaswag|10_2023-09-03T12:27:36.436118.parquet
  • harness_hendrycksTest_5

    • 数据文件路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-management|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-professional_law|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-professional_medicine|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-professional_psychology|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-public_relations|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-security_studies|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-sociology|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-us_foreign_policy|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-virology|5_2023-09-03T12:27:36.436118.parquet
      • **/details_harness|hendrycksTest-world_religions|5_2023-09-03T12:27:36.436118.parquet }
  • harness_hendrycksTest_abstract_algebra_5

    • 数据文件路径:
      • `**/details_
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