five

open-llm-leaderboard-old/details_ICBU-NPU__FashionGPT-70B-V1

收藏
Hugging Face2023-10-29 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_ICBU-NPU__FashionGPT-70B-V1
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ICBU-NPU/FashionGPT-70B-V1时自动创建的,用于Open LLM Leaderboard的评估。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,每次运行的结果作为一个特定的分割存储,分割名称使用运行的时间戳。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。

该数据集是在评估模型ICBU-NPU/FashionGPT-70B-V1时自动创建的,用于Open LLM Leaderboard的评估。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,每次运行的结果作为一个特定的分割存储,分割名称使用运行的时间戳。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。
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 ICBU-NPU/FashionGPT-70B-V1 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_ICBU-NPU__FashionGPT-70B-V1", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-29T15:24:49.736716 运行的最新结果

python { "all": { "em": 0.0019924496644295304, "em_stderr": 0.0004566676462666993, "f1": 0.07125104865771813, "f1_stderr": 0.0014102826102321945, "acc": 0.5589478168926856, "acc_stderr": 0.011387742640607 }, "harness|drop|3": { "em": 0.0019924496644295304, "em_stderr": 0.0004566676462666993, "f1": 0.07125104865771813, "f1_stderr": 0.0014102826102321945 }, "harness|gsm8k|5": { "acc": 0.2812736921910538, "acc_stderr": 0.012384789310940239 }, "harness|winogrande|5": { "acc": 0.8366219415943172, "acc_stderr": 0.010390695970273764 } }

配置详情

以下是数据集的部分配置详情:

  • harness_arc_challenge_25

    • 分割: 2023_09_19T01_12_17.792946
    • 路径: **/details_harness|arc:challenge|25_2023-09-19T01-12-17.792946.parquet
    • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2023-09-19T01-12-17.792946.parquet
  • harness_drop_3

    • 分割: 2023_10_29T15_24_49.736716
    • 路径: **/details_harness|drop|3_2023-10-29T15-24-49.736716.parquet
    • 分割: latest
    • 路径: **/details_harness|drop|3_2023-10-29T15-24-49.736716.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_29T15_24_49.736716
    • 路径: **/details_harness|gsm8k|5_2023-10-29T15-24-49.736716.parquet
    • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2023-10-29T15-24-49.736716.parquet
  • harness_hellaswag_10

    • 分割: 2023_09_19T01_12_17.792946
    • 路径: **/details_harness|hellaswag|10_2023-09-19T01-12-17.792946.parquet
    • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2023-09-19T01-12-17.792946.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_09_19T01_12_17.792946
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-management|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-09-19T01-12-17.792946.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-19T01-12-17.792946.parquet
      • `**/details_harness|hendrycksTest-professional_law|5_2023-09-19T01-12-17.7929
二维码
社区交流群
二维码
科研交流群
商业服务