five

open-llm-leaderboard/details_Severian__ANIMA-Phi-Neptune-Mistral-7B

收藏
Hugging Face2023-10-29 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_Severian__ANIMA-Phi-Neptune-Mistral-7B
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Severian/ANIMA-Phi-Neptune-Mistral-7B在Open LLM排行榜上的评估运行期间自动创建的。数据集包含64个配置,每个配置对应一个评估任务。它包含来自2次运行的结果,每次运行都以使用运行时间戳命名的特定分割表示。train分割始终指向最新的结果。额外的results配置存储了运行中的所有聚合结果,用于在排行榜上计算和显示聚合指标。数据集结构包含多个配置,每个配置详细说明不同的任务及其相应的结果。

该数据集是在模型Severian/ANIMA-Phi-Neptune-Mistral-7B在Open LLM排行榜上的评估运行期间自动创建的。数据集包含64个配置,每个配置对应一个评估任务。它包含来自2次运行的结果,每次运行都以使用运行时间戳命名的特定分割表示。train分割始终指向最新的结果。额外的results配置存储了运行中的所有聚合结果,用于在排行榜上计算和显示聚合指标。数据集结构包含多个配置,每个配置详细说明不同的任务及其相应的结果。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

该数据集是在评估模型 Severian/ANIMA-Phi-Neptune-Mistral-7BOpen LLM Leaderboard 上的运行过程中自动创建的。

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。

最新结果

以下是 2023-10-29T10:04:15.191273 运行的最新结果

python { "all": { "em": 0.10329278523489933, "em_stderr": 0.003116735713102519, "f1": 0.1624748322147643, "f1_stderr": 0.003266242273162539, "acc": 0.442081101118795, "acc_stderr": 0.011112320094960076 }, "harness|drop|3": { "em": 0.10329278523489933, "em_stderr": 0.003116735713102519, "f1": 0.1624748322147643, "f1_stderr": 0.003266242273162539 }, "harness|gsm8k|5": { "acc": 0.14935557240333586, "acc_stderr": 0.009818090723727293 }, "harness|winogrande|5": { "acc": 0.7348066298342542, "acc_stderr": 0.01240654946619286 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割: 2023_10_10T23_09_12.843992
      • 路径: **/details_harness|arc:challenge|25_2023-10-10T23-09-12.843992.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2023-10-10T23-09-12.843992.parquet
  • harness_drop_3

    • 分割: 2023_10_29T10_04_15.191273
      • 路径: **/details_harness|drop|3_2023-10-29T10-04-15.191273.parquet
    • 分割: latest
      • 路径: **/details_harness|drop|3_2023-10-29T10-04-15.191273.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_29T10_04_15.191273
      • 路径: **/details_harness|gsm8k|5_2023-10-29T10-04-15.191273.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2023-10-29T10-04-15.191273.parquet
  • harness_hellaswag_10

    • 分割: 2023_10_10T23_09_12.843992
      • 路径: **/details_harness|hellaswag|10_2023-10-10T23-09-12.843992.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2023-10-10T23-09-12.843992.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_10_10T23_09_12.843992
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-management|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-10-10T23-09-12.843992.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-10-10T23-09-12.843992.parquet
        • `**/details_harness|hendrycksTest-professional_psychology|5
二维码
社区交流群
二维码
科研交流群
商业服务