five

open-llm-leaderboard-old/details_openaccess-ai-collective__minotaur-13b-fixed

收藏
Hugging Face2023-09-17 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_openaccess-ai-collective__minotaur-13b-fixed
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型 openaccess-ai-collective/minotaur-13b-fixed 在 Open LLM Leaderboard 上的评估运行过程中自动生成的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。它由 2 次运行创建,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。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。README 还提供了如何使用 Python 代码加载运行中的详细信息的示例,并包含了特定运行的最新结果。

该数据集是在模型 openaccess-ai-collective/minotaur-13b-fixed 在 Open LLM Leaderboard 上的评估运行过程中自动生成的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。它由 2 次运行创建,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。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。README 还提供了如何使用 Python 代码加载运行中的详细信息的示例,并包含了特定运行的最新结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

该数据集是在对模型 openaccess-ai-collective/minotaur-13b-fixed 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集组成

数据集由64个配置组成,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。

结果配置

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。

加载数据示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_openaccess-ai-collective__minotaur-13b-fixed", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-17T11:35:58.500746 运行的最新结果

python { "all": { "em": 0.08588506711409397, "em_stderr": 0.0028694514614574086, "f1": 0.15832529362416004, "f1_stderr": 0.0031260511951114243, "acc": 0.44995251838080197, "acc_stderr": 0.01057426968021918 }, "harness|drop|3": { "em": 0.08588506711409397, "em_stderr": 0.0028694514614574086, "f1": 0.15832529362416004, "f1_stderr": 0.0031260511951114243 }, "harness|gsm8k|5": { "acc": 0.13115996967399546, "acc_stderr": 0.009298499235587853 }, "harness|winogrande|5": { "acc": 0.7687450670876085, "acc_stderr": 0.01185004012485051 } }

配置详情

  • harness_arc_challenge_25

    • 分割: 2023_07_24T12_56_58.097671
    • 路径: **/details_harness|arc:challenge|25_2023-07-24T12:56:58.097671.parquet
    • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2023-07-24T12:56:58.097671.parquet
  • harness_drop_3

    • 分割: 2023_09_17T11_35_58.500746
    • 路径: **/details_harness|drop|3_2023-09-17T11-35-58.500746.parquet
    • 分割: latest
    • 路径: **/details_harness|drop|3_2023-09-17T11-35-58.500746.parquet
  • harness_gsm8k_5

    • 分割: 2023_09_17T11_35_58.500746
    • 路径: **/details_harness|gsm8k|5_2023-09-17T11-35-58.500746.parquet
    • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2023-09-17T11-35-58.500746.parquet
  • harness_hellaswag_10

    • 分割: 2023_07_24T12_56_58.097671
    • 路径: **/details_harness|hellaswag|10_2023-07-24T12:56:58.097671.parquet
    • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2023-07-24T12:56:58.097671.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_07_24T12_56_58.097671
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-management|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-24T12:56:58.097671.parquet
      • **/details_harness|hendrycksTest-professional_law|5_2023-07-24T12:56:58.097671.parquet
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