five

open-llm-leaderboard/details_sauce1337__AppleSauce-L2-13b

收藏
Hugging Face2023-10-23 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_sauce1337__AppleSauce-L2-13b
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sauce1337/AppleSauce-L2-13b在Open LLM Leaderboard上的评估运行期间自动创建的。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次运行生成,每次运行的结果作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,并用于在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

该数据集是在评估模型 sauce1337/AppleSauce-L2-13bOpen LLM Leaderboard 上的自动创建的。数据集包含64个配置,每个配置对应一个评估任务。

数据集结构

  • 配置数量:64个配置
  • 数据来源:从2次运行中创建,每次运行对应一个特定的时间戳。
  • 数据分割:每个配置包含多个分割,其中“train”分割指向最新的结果。
  • 结果汇总:额外配置“results”存储所有运行的汇总结果,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_sauce1337__AppleSauce-L2-13b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-23T06:05:52.891259 运行 的最新结果:

python { "all": { "em": 0.3851719798657718, "em_stderr": 0.004983607378769111, "f1": 0.4518634647651026, "f1_stderr": 0.004795271144996667, "acc": 0.4296748451536735, "acc_stderr": 0.010140916870454906 }, "harness|drop|3": { "em": 0.3851719798657718, "em_stderr": 0.004983607378769111, "f1": 0.4518634647651026, "f1_stderr": 0.004795271144996667 }, "harness|gsm8k|5": { "acc": 0.10007581501137225, "acc_stderr": 0.008266274528685634 }, "harness|winogrande|5": { "acc": 0.7592738752959748, "acc_stderr": 0.01201555921222418 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|arc:challenge|25_2023-09-13T16-32-51.732119.parquet
  • harness_drop_3

    • 分割:2023_10_23T06_05_52.891259, latest
    • 路径:**/details_harness|drop|3_2023-10-23T06-05-52.891259.parquet
  • harness_gsm8k_5

    • 分割:2023_10_23T06_05_52.891259, latest
    • 路径:**/details_harness|gsm8k|5_2023-10-23T06-05-52.891259.parquet
  • harness_hellaswag_10

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hellaswag|10_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:多个路径,例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-13T16-32-51.732119.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割:2023_09_13T16_32_51.732119, latest
    • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-13T16-32-51.732119.parquet

以上是数据集的详细概述,包括数据集的结构、加载示例、最新结果以及各个配置的详细信息。

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