five

open-llm-leaderboard/details_IkariDev__Athena-v1

收藏
Hugging Face2023-10-28 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_IkariDev__Athena-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Open LLM Leaderboard上对模型IkariDev/Athena-v1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,每次运行都作为一个特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新的结果。此外,还有一个results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。

This dataset was automatically generated during the evaluation of the model IkariDev/Athena-v1 on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to a single evaluation task. The dataset is created from two experimental runs, with each run serving as a dedicated data split whose name is derived from the timestamp of the corresponding run. The `train` split always references the most recent evaluation results. Additionally, a `results` configuration is included to store the aggregated results across all runs, which is utilized for computing and displaying the aggregate metrics on the Open LLM Leaderboard.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

该数据集是在对模型 IkariDev/Athena-v1 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。
  • 一个额外的配置 "results" 存储所有聚合的运行结果,用于计算和显示 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_IkariDev__Athena-v1", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-28T04:34:50.087608 运行 的最新结果:

python { "all": { "em": 0.369756711409396, "em_stderr": 0.004943696476134741, "f1": 0.44112835570469994, "f1_stderr": 0.004761323223659711, "acc": 0.39875195446551615, "acc_stderr": 0.00908031987803613 }, "harness|drop|3": { "em": 0.369756711409396, "em_stderr": 0.004943696476134741, "f1": 0.44112835570469994, "f1_stderr": 0.004761323223659711 }, "harness|gsm8k|5": { "acc": 0.04927975739196361, "acc_stderr": 0.005962150655812479 }, "harness|winogrande|5": { "acc": 0.7482241515390686, "acc_stderr": 0.012198489100259781 } }

配置详情

  • harness_arc_challenge_25

    • 分割: 2023_09_15T01_58_36.920407
    • 路径: **/details_harness|arc:challenge|25_2023-09-15T01-58-36.920407.parquet
    • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2023-09-15T01-58-36.920407.parquet
  • harness_drop_3

    • 分割: 2023_10_28T04_34_50.087608
    • 路径: **/details_harness|drop|3_2023-10-28T04-34-50.087608.parquet
    • 分割: latest
    • 路径: **/details_harness|drop|3_2023-10-28T04-34-50.087608.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_28T04_34_50.087608
    • 路径: **/details_harness|gsm8k|5_2023-10-28T04-34-50.087608.parquet
    • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2023-10-28T04-34-50.087608.parquet
  • harness_hellaswag_10

    • 分割: 2023_09_15T01_58_36.920407
    • 路径: **/details_harness|hellaswag|10_2023-09-15T01-58-36.920407.parquet
    • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2023-09-15T01-58-36.920407.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_09_15T01_58_36.920407
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-management|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-15T01-58-36.920407.parquet
      • **/details_harness|hendrycksTest-professional_law|5_2023-09-15T01-58-36.920407.parquet
      • `**/details_harness|hendrycksTest-professional_medicine|5_2023-09-15T0
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