five

open-llm-leaderboard/details_KnutJaegersberg__black_goo_recipe_d

收藏
Hugging Face2023-10-16 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_KnutJaegersberg__black_goo_recipe_d
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KnutJaegersberg/black_goo_recipe_d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次运行创建,每次运行在每个配置中作为一个特定的分割存在。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。

This dataset was automatically generated when evaluating the model KnutJaegersberg/black_goo_recipe_d on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to a single evaluation task. The dataset is built from two runs, where each run exists as a specific split within every configuration. The `train` split always points to the most recent results. In addition, there is a configuration named `results` that stores the aggregated results of all runs, which is utilized to compute and display the aggregated metrics on the Open LLM Leaderboard. The README also provides examples of how to load run details using the `datasets` library in Python.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集来源

该数据集是在评估模型 KnutJaegersberg/black_goo_recipe_dOpen LLM Leaderboard 上的自动创建的。

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" 存储了所有运行结果的聚合,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_KnutJaegersberg__black_goo_recipe_d", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果 来自 2023-10-16T10:23:29.874425 运行的结果: python { "all": { "em": 0.0008389261744966443, "em_stderr": 0.00029649629898012315, "f1": 0.05404362416107403, "f1_stderr": 0.0013498976751710832, "acc": 0.31996956628667406, "acc_stderr": 0.007607094818516801 }, "harness|drop|3": { "em": 0.0008389261744966443, "em_stderr": 0.00029649629898012315, "f1": 0.05404362416107403, "f1_stderr": 0.0013498976751710832 }, "harness|gsm8k|5": { "acc": 0.0037907505686125853, "acc_stderr": 0.0016927007401501906 }, "harness|winogrande|5": { "acc": 0.6361483820047356, "acc_stderr": 0.013521488896883411 } }

配置详情

以下是数据集的配置详情:

配置列表

  • harness_arc_challenge_25

    • 分割: 2023_09_03T11_42_32.382522
      • 路径: **/details_harness|arc:challenge|25_2023-09-03T11:42:32.382522.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2023-09-03T11:42:32.382522.parquet
  • harness_drop_3

    • 分割: 2023_10_16T10_23_29.874425
      • 路径: **/details_harness|drop|3_2023-10-16T10-23-29.874425.parquet
    • 分割: latest
      • 路径: **/details_harness|drop|3_2023-10-16T10-23-29.874425.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_16T10_23_29.874425
      • 路径: **/details_harness|gsm8k|5_2023-10-16T10-23-29.874425.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2023-10-16T10-23-29.874425.parquet
  • harness_hellaswag_10

    • 分割: 2023_09_03T11_42_32.382522
      • 路径: **/details_harness|hellaswag|10_2023-09-03T11:42:32.382522.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2023-09-03T11:42:32.382522.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_09_03T11_42_32.382522
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-management|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-09-03T11:42:32.382522.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-03T11:42:32.382522.parquet
        • `**/details_harness|hendrycksTest-professional_law|5_2023-09
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