five

open-llm-leaderboard-old/details_Undi95__Nous-Hermes-13B-Code

收藏
Hugging Face2023-10-17 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Undi95__Nous-Hermes-13B-Code
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Undi95/Nous-Hermes-13B-Code在Open LLM Leaderboard上进行评估时自动生成的。数据集由64个配置组成,每个配置对应一个被评估的任务。它包含2次运行的数据,每次运行在每个配置中表示为特定的分割,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。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了如何使用Python代码加载运行中的详细信息的示例,并包含了特定运行的最新结果。

该数据集是在模型Undi95/Nous-Hermes-13B-Code在Open LLM Leaderboard上进行评估时自动生成的。数据集由64个配置组成,每个配置对应一个被评估的任务。它包含2次运行的数据,每次运行在每个配置中表示为特定的分割,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。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。README还提供了如何使用Python代码加载运行中的详细信息的示例,并包含了特定运行的最新结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 Undi95/Nous-Hermes-13B-CodeOpen 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 次运行结果组成,每个运行结果作为一个特定的分割(split)存储在每个配置中,分割名称使用运行的时间戳。"train" 分割始终指向最新的结果。

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

以下是加载特定运行详细信息的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Undi95__Nous-Hermes-13B-Code", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-17T01:46:49.269980 运行的最新结果: python { "all": { "em": 0.19043624161073824, "em_stderr": 0.004021054701391535, "f1": 0.28277894295302086, "f1_stderr": 0.004086388636430754, "acc": 0.42762389052479904, "acc_stderr": 0.010275468471163573 }, "harness|drop|3": { "em": 0.19043624161073824, "em_stderr": 0.004021054701391535, "f1": 0.28277894295302086, "f1_stderr": 0.004086388636430754 }, "harness|gsm8k|5": { "acc": 0.10386656557998483, "acc_stderr": 0.008403622228924035 }, "harness|winogrande|5": { "acc": 0.7513812154696132, "acc_stderr": 0.012147314713403108 } }

配置详情

以下是数据集的配置详情:

  • config_name: harness_arc_challenge_25

    • split: 2023_09_05T02_42_01.860222
      • path: **/details_harness|arc:challenge|25_2023-09-05T02:42:01.860222.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-09-05T02:42:01.860222.parquet
  • config_name: harness_drop_3

    • split: 2023_10_17T01_46_49.269980
      • path: **/details_harness|drop|3_2023-10-17T01-46-49.269980.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-10-17T01-46-49.269980.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_10_17T01_46_49.269980
      • path: **/details_harness|gsm8k|5_2023-10-17T01-46-49.269980.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-10-17T01-46-49.269980.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_09_05T02_42_01.860222
      • path: **/details_harness|hellaswag|10_2023-09-05T02:42:01.860222.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-09-05T02:42:01.860222.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_09_05T02_42_01.860222
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-management|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-05T02:42:01.860222.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-09-05T
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