five

open-llm-leaderboard/details_bhenrym14__airophin-13b-pntk-16k-fp16

收藏
Hugging Face2023-09-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_bhenrym14__airophin-13b-pntk-16k-fp16
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型bhenrym14/airophin-13b-pntk-16k-fp16时自动创建的,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,每次运行的结果存储在不同的分割中,最新的结果存储在train分割中。此外,数据集还包含一个名为results的配置,用于存储所有运行的聚合结果。

该数据集是在评估模型bhenrym14/airophin-13b-pntk-16k-fp16时自动创建的,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,每次运行的结果存储在不同的分割中,最新的结果存储在train分割中。此外,数据集还包含一个名为results的配置,用于存储所有运行的聚合结果。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 bhenrym14/airophin-13b-pntk-16k-fp16Open 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作为一个特定的分片找到,分片名称使用运行的时间戳。
  • "train" 分片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。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_bhenrym14__airophin-13b-pntk-16k-fp16", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-22T13:50:50.012213 运行的最新结果: python { "all": { "em": 0.12793624161073824, "em_stderr": 0.003420665162123249, "f1": 0.20498217281879216, "f1_stderr": 0.0035591754960457413, "acc": 0.421002792649235, "acc_stderr": 0.009731603620470694 }, "harness|drop|3": { "em": 0.12793624161073824, "em_stderr": 0.003420665162123249, "f1": 0.20498217281879216, "f1_stderr": 0.0035591754960457413 }, "harness|gsm8k|5": { "acc": 0.0803639120545868, "acc_stderr": 0.007488258573239077 }, "harness|winogrande|5": { "acc": 0.7616416732438832, "acc_stderr": 0.01197494866770231 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|arc:challenge|25_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|arc:challenge|25_2023-08-09T13:13:26.207427.parquet
  • harness_drop_3

    • 分片:2023_09_22T13_50_50.012213
      • 路径:**/details_harness|drop|3_2023-09-22T13-50-50.012213.parquet
    • 分片:latest
      • 路径:**/details_harness|drop|3_2023-09-22T13-50-50.012213.parquet
  • harness_gsm8k_5

    • 分片:2023_09_22T13_50_50.012213
      • 路径:**/details_harness|gsm8k|5_2023-09-22T13-50-50.012213.parquet
    • 分片:latest
      • 路径:**/details_harness|gsm8k|5_2023-09-22T13-50-50.012213.parquet
  • harness_hellaswag_10

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hellaswag|10_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hellaswag|10_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_psychology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-public_relations|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-security_studies|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-sociology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-us_foreign_policy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-virology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-world_religions|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-professional_psychology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-public_relations|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-security_studies|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-sociology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-us_foreign_policy|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-virology|5_2023-08-09T13:13:26.207427.parquet
        • **/details_harness|hendrycksTest-world_religions|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_anatomy_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_astronomy_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_college_biology_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_college_physics_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_computer_security_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_econometrics_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分片:2023_08_09T13_13_26.207427
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T13:13:26.207427.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分片:2023_08_09T13_13_26
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T13:13:26.207427.parquet
    • 分片:latest
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T13:13:26.207427.parquet
二维码
社区交流群
二维码
科研交流群
商业服务