five

open-llm-leaderboard/details_JosephusCheung__LL7M

收藏
Hugging Face2023-10-24 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_JosephusCheung__LL7M
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JosephusCheung/LL7M进行评估时自动创建的。它由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,每次运行在每个配置中表示为特定的分割。train分割始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。

This dataset was automatically created during the evaluation of the model JosephusCheung/LL7M on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to an evaluation task. The dataset is generated from two runs, where each run is denoted as a specific split within each configuration. The `train` split always points to the most recent results. An additional configuration named `results` stores the aggregated results across all runs, which are used to calculate and display the aggregate metrics on the Open LLM Leaderboard. This dataset can be loaded using the `load_dataset` function from the `datasets` library.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集来源

数据集结构

  • 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。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_JosephusCheung__LL7M", "harness_winogrande_5", split="train")

最新结果

  • 这些是最新结果,来自 2023-10-24T03:13:24.379539 的运行: python { "all": { "em": 0.019819630872483222, "em_stderr": 0.0014273827117586067, "f1": 0.07556312919463118, "f1_stderr": 0.0018868261588306972, "acc": 0.3234745894051663, "acc_stderr": 0.007810892751790338 }, "harness|drop|3": { "em": 0.019819630872483222, "em_stderr": 0.0014273827117586067, "f1": 0.07556312919463118, "f1_stderr": 0.0018868261588306972 }, "harness|gsm8k|5": { "acc": 0.006065200909780136, "acc_stderr": 0.00213867030146044 }, "harness|winogrande|5": { "acc": 0.6408839779005525, "acc_stderr": 0.013483115202120236 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|arc:challenge|25_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-10-10T15-26-54.562937.parquet
  • harness_drop_3

    • 分割:2023_10_24T03_13_24.379539
    • 路径:**/details_harness|drop|3_2023-10-24T03-13-24.379539.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-10-24T03-13-24.379539.parquet
  • harness_gsm8k_5

    • 分割:2023_10_24T03_13_24.379539
    • 路径:**/details_harness|gsm8k|5_2023-10-24T03-13-24.379539.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-24T03-13-24.379539.parquet
  • harness_hellaswag_10

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hellaswag|10_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T15-26-54.562937.parquet 等 42 个文件
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T15-26-54.562937.parquet 等 42 个文件
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_10_10T15_26_54.562937
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T15-26-54.562937.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T15-26-54.562937.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割:2023_10_10
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