five

open-llm-leaderboard-old/details_KnutJaegersberg__black_goo_recipe_b

收藏
Hugging Face2023-10-17 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_KnutJaegersberg__black_goo_recipe_b
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_b进行评估时自动生成的。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,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,名为results的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。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Open LLM Leaderboard上对模型KnutJaegersberg/black_goo_recipe_b进行评估时自动生成的。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,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,名为results的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。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

该数据集是在模型 KnutJaegersberg/black_goo_recipe_bOpen LLM Leaderboard 上的评估运行期间自动创建的。

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。
  • 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。

加载数据集示例

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

最新结果

以下是 2023-10-17T14:56:50.691599 运行的最新结果

python { "all": { "em": 0.0007340604026845638, "em_stderr": 0.0002773614457335715, "f1": 0.05097630033557055, "f1_stderr": 0.0013271541576312406, "acc": 0.3192425320418652, "acc_stderr": 0.007133502794987516 }, "harness|drop|3": { "em": 0.0007340604026845638, "em_stderr": 0.0002773614457335715, "f1": 0.05097630033557055, "f1_stderr": 0.0013271541576312406 }, "harness|gsm8k|5": { "acc": 0.000758150113722517, "acc_stderr": 0.0007581501137225241 }, "harness|winogrande|5": { "acc": 0.6377269139700079, "acc_stderr": 0.013508855476252508 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_08_31T14_15_51.764812
      • 路径:**/details_harness|arc:challenge|25_2023-08-31T14:15:51.764812.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-08-31T14:15:51.764812.parquet
  • harness_drop_3

    • 分割:2023_10_17T14_56_50.691599
      • 路径:**/details_harness|drop|3_2023-10-17T14-56-50.691599.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-17T14-56-50.691599.parquet
  • harness_gsm8k_5

    • 分割:2023_10_17T14_56_50.691599
      • 路径:**/details_harness|gsm8k|5_2023-10-17T14-56-50.691599.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-17T14-56-50.691599.parquet
  • harness_hellaswag_10

    • 分割:2023_08_31T14_15_51.764812
      • 路径:**/details_harness|hellaswag|10_2023-08-31T14:15:51.764812.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-08-31T14:15:51.764812.parquet
  • harness_hendrycksTest_5

    • 分割:2023_08_31T14_15_51.764812
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-31T14:15:51.764812.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-08-31T14:15:51.764812.parquet
二维码
社区交流群
二维码
科研交流群
商业服务