five

open-llm-leaderboard-old/details_yeen214__llama2_7b_small_tuning_v1

收藏
Hugging Face2023-10-23 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_yeen214__llama2_7b_small_tuning_v1
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型yeen214/llama2_7b_small_tuning_v1的评估过程中自动创建的。数据集由64个配置组成,每个配置对应一个评估任务。数据集由2次运行创建,每次运行的结果都存储为特定配置中的一个分割,分割名称使用运行的时间戳。"train"分割始终指向最新的结果。此外,还有一个名为"results"的配置,存储了所有运行的聚合结果,用于计算和展示在Open LLM Leaderboard上的聚合指标。

该数据集是在模型yeen214/llama2_7b_small_tuning_v1的评估过程中自动创建的。数据集由64个配置组成,每个配置对应一个评估任务。数据集由2次运行创建,每次运行的结果都存储为特定配置中的一个分割,分割名称使用运行的时间戳。"train"分割始终指向最新的结果。此外,还有一个名为"results"的配置,存储了所有运行的聚合结果,用于计算和展示在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

该数据集是在评估模型 yeen214/llama2_7b_small_tuning_v1Open LLM Leaderboard 上的运行过程中自动创建的。

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一个特定的拆分,拆分名称使用运行的时间戳。"train" 拆分始终指向最新的结果。

额外配置

一个额外的配置 "results" 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。

加载数据示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_yeen214__llama2_7b_small_tuning_v1", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果来自 run 2023-10-23T22:59:12.281611 的示例:

python { "all": { "em": 0.0, "em_stderr": 0.0, "f1": 5.872483221476512e-05, "f1_stderr": 1.3579559736417577e-05, "acc": 0.24861878453038674, "acc_stderr": 0.007026135605808226 }, "harness|drop|3": { "em": 0.0, "em_stderr": 0.0, "f1": 5.872483221476512e-05, "f1_stderr": 1.3579559736417577e-05 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|winogrande|5": { "acc": 0.4972375690607735, "acc_stderr": 0.014052271211616452 } }

配置详情

配置列表

  • harness_arc_challenge_25

    • 拆分: 2023_10_04T06_48_03.956083
      • 路径: **/details_harness|arc:challenge|25_2023-10-04T06-48-03.956083.parquet
    • 拆分: latest
      • 路径: **/details_harness|arc:challenge|25_2023-10-04T06-48-03.956083.parquet
  • harness_drop_3

    • 拆分: 2023_10_23T22_59_12.281611
      • 路径: **/details_harness|drop|3_2023-10-23T22-59-2.281611.parquet
    • 拆分: latest
      • 路径: **/details_harness|drop|3_2023-10-23T22-59-2.281611.parquet
  • harness_gsm8k_5

    • 拆分: 2023_10_23T22_59_12.281611
      • 路径: **/details_harness|gsm8k|5_2023-10-23T22-59-2.281611.parquet
    • 拆分: latest
      • 路径: **/details_harness|gsm8k|5_2023-10-23T22-59-2.281611.parquet
  • harness_hellaswag_10

    • 拆分: 2023_10_04T06_48_03.956083
      • 路径: **/details_harness|hellaswag|10_2023-10-04T06-48-03.956083.parquet
    • 拆分: latest
      • 路径: **/details_harness|hellaswag|10_2023-10-04T06-48-03.956083.parquet
  • harness_hendrycksTest_5

    • 拆分: 2023_10_04T06_48_03.956083
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-management|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-10-04T06-48-03.956083.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-10-04T06-48-03.956083.parquet
        • `**/details_harness|hendrycksTest-
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