five

open-llm-leaderboard/details_Xilabs__calypso-3b-alpha-v2

收藏
Hugging Face2023-10-24 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_Xilabs__calypso-3b-alpha-v2
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型Xilabs/calypso-3b-alpha-v2进行评估时自动生成的。数据集包含64个配置,每个配置对应一个评估任务。数据集由2次运行生成,每次运行在每个配置中作为一个特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新的结果。此外,还有一个results配置,用于存储所有运行的聚合结果,并用于计算和显示在Open LLM Leaderboard上的聚合指标。README还提供了如何使用Hugging Face datasets库加载数据集的示例。
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 Xilabs/calypso-3b-alpha-v2Open 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

数据集由 2 次运行创建,每次运行可以在每个配置中找到特定的拆分,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。

加载数据示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Xilabs__calypso-3b-alpha-v2", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-24T19:29:31.085880 运行的最新结果

python { "all": { "em": 0.17051174496644295, "em_stderr": 0.0038514292227271175, "f1": 0.2208368288590607, "f1_stderr": 0.0039617614020142555, "acc": 0.3297731593317987, "acc_stderr": 0.007824223176002862 }, "harness|drop|3": { "em": 0.17051174496644295, "em_stderr": 0.0038514292227271175, "f1": 0.2208368288590607, "f1_stderr": 0.0039617614020142555 }, "harness|gsm8k|5": { "acc": 0.006823351023502654, "acc_stderr": 0.0022675371022544896 }, "harness|winogrande|5": { "acc": 0.6527229676400947, "acc_stderr": 0.013380909249751233 } }

配置详情

配置列表

  • harness_arc_challenge_25

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|arc:challenge|25_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|arc:challenge|25_2023-10-03T14-01-45.504923.parquet
  • harness_drop_3

    • 拆分:2023_10_24T19_29_31.085880
    • 路径:**/details_harness|drop|3_2023-10-24T19-29-31.085880.parquet
    • 拆分:latest
    • 路径:**/details_harness|drop|3_2023-10-24T19-29-31.085880.parquet
  • harness_gsm8k_5

    • 拆分:2023_10_24T19_29_31.085880
    • 路径:**/details_harness|gsm8k|5_2023-10-24T19-29-31.085880.parquet
    • 拆分:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-24T19-29-31.085880.parquet
  • harness_hellaswag_10

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hellaswag|10_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hellaswag|10_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T14-01-45.504923.parquet 等 40 个文件
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T14-01-45.504923.parquet 等 40 个文件
  • harness_hendrycksTest_abstract_algebra_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_anatomy_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_astronomy_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_business_ethics_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_college_biology_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_college_medicine_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_college_physics_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_computer_security_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_econometrics_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-03T14-01-45.504923.parquet
    • 拆分:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-03T14-01-45.504923.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 拆分:2023_10_03T14_01_45.504923
    • 路径:`**/details_harness|hendrycksTest-electrical_engineering|
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