five

open-llm-leaderboard/details_jondurbin__airoboros-13b-gpt4-1.3

收藏
Hugging Face2023-10-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_jondurbin__airoboros-13b-gpt4-1.3
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jondurbin/airoboros-13b-gpt4-1.3时自动生成的,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,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,用于在Open LLM Leaderboard上计算和显示聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集是在评估模型 jondurbin/airoboros-13b-gpt4-1.3Open 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。每个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。

额外配置

  • 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示在 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_jondurbin__airoboros-13b-gpt4-1.3", "harness_winogrande_5", split="train")

最新结果

这些是最新结果,来自 2023-10-22T11:55:30.055248 的运行: python { "all": { "em": 0.018036912751677854, "em_stderr": 0.0013629136303228225, "f1": 0.08087038590604041, "f1_stderr": 0.0018769490051502272, "acc": 0.3909936314193683, "acc_stderr": 0.008100933725827915 }, "harness|drop|3": { "em": 0.018036912751677854, "em_stderr": 0.0013629136303228225, "f1": 0.08087038590604041, "f1_stderr": 0.0018769490051502272 }, "harness|gsm8k|5": { "acc": 0.02350265352539803, "acc_stderr": 0.004172883669643956 }, "harness|winogrande|5": { "acc": 0.7584846093133386, "acc_stderr": 0.012028983782011874 } }

配置详情

  • harness_arc_challenge_25

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|arc:challenge|25_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|arc:challenge|25_2023-08-09T08:50:11.313288.parquet
  • harness_drop_3

    • 分片: 2023_10_22T11_55_30.055248
    • 路径: **/details_harness|drop|3_2023-10-22T11-55-30.055248.parquet
    • 分片: latest
    • 路径: **/details_harness|drop|3_2023-10-22T11-55-30.055248.parquet
  • harness_gsm8k_5

    • 分片: 2023_10_22T11_55_30.055248
    • 路径: **/details_harness|gsm8k|5_2023-10-22T11-55-30.055248.parquet
    • 分片: latest
    • 路径: **/details_harness|gsm8k|5_2023-10-22T11-55-30.055248.parquet
  • harness_hellaswag_10

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hellaswag|10_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hellaswag|10_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T08:50:11.313288.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-08-09T08:50:11.313288.parquet
      • ...(其他路径省略)
    • 分片: latest
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T08:50:11.313288.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-08-09T08:50:11.313288.parquet
      • ...(其他路径省略)
  • harness_hendrycksTest_abstract_algebra_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_anatomy_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_astronomy_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_college_biology_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_college_physics_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_computer_security_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T08:50:11.313288.parquet
  • harness_hendrycksTest_econometrics_5

    • 分片: 2023_08_09T08_50_11.313288
    • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-08-09T08:50:11.313288.parquet
    • 分片: latest
    • 路径: `**/details_harness|hendrycksTest-econometrics|5_2023-08-09T08:50:11.313288.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