five

open-llm-leaderboard/details_TinyPixel__lima-test

收藏
Hugging Face2023-10-17 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_TinyPixel__lima-test
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TinyPixel/lima-test的评估运行期间自动创建的,用于Open LLM Leaderboard的评估。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split,split以运行的时间戳命名。train split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
原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 TinyPixel/lima-test 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集结构

  • 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。
  • 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 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_TinyPixel__lima-test", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-17T19:33:22.756804 运行的最新结果: python { "all": { "em": 0.001363255033557047, "em_stderr": 0.0003778609196461008, "f1": 0.05645763422818797, "f1_stderr": 0.0013211879752480866, "acc": 0.40996862682678575, "acc_stderr": 0.009889312522114477 }, "harness|drop|3": { "em": 0.001363255033557047, "em_stderr": 0.0003778609196461008, "f1": 0.05645763422818797, "f1_stderr": 0.0013211879752480866 }, "harness|gsm8k|5": { "acc": 0.07960576194086429, "acc_stderr": 0.007455924338676286 }, "harness|winogrande|5": { "acc": 0.7403314917127072, "acc_stderr": 0.012322700705552667 } }

配置详情

以下是数据集的配置详情:

配置列表

  • harness_arc_challenge_25

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|arc:challenge|25_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|arc:challenge|25_2023-08-28T09:10:45.645303.parquet
  • harness_drop_3

    • 分片:2023_10_17T19_33_22.756804
    • 路径:**/details_harness|drop|3_2023-10-17T19-33-22.756804.parquet
    • 分片:latest
    • 路径:**/details_harness|drop|3_2023-10-17T19-33-22.756804.parquet
  • harness_gsm8k_5

    • 分片:2023_10_17T19_33_22.756804
    • 路径:**/details_harness|gsm8k|5_2023-10-17T19-33-22.756804.parquet
    • 分片:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-17T19-33-22.756804.parquet
  • harness_hellaswag_10

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hellaswag|10_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hellaswag|10_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_anatomy_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_astronomy_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_college_biology_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_college_physics_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_computer_security_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_econometrics_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-28T09:10:45.645303.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-28T09:10:45.645303.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分片:2023_08_28T09_10_45.645303
    • 路径:`**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-28T09:10:45.645303.
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