five

open-llm-leaderboard-old/details_TheBloke__Orca-2-13B-GPTQ

收藏
Hugging Face2023-11-25 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_TheBloke__Orca-2-13B-GPTQ
下载链接
链接失效反馈
官方服务:
资源简介:
数据集是在模型TheBloke/Orca-2-13B-GPTQ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。数据集从1次运行中创建,每次运行可以在每个配置中找到特定的分割,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,results配置存储了运行的所有聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。

数据集是在模型TheBloke/Orca-2-13B-GPTQ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。数据集从1次运行中创建,每次运行可以在每个配置中找到特定的分割,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,results配置存储了运行的所有聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集描述

该数据集是在对模型 TheBloke/Orca-2-13B-GPTQ 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

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 1 次运行中创建,每个运行可以在每个配置中找到特定的分割,分割名称使用运行的时间戳。
  • "train" 分割始终指向最新的结果。
  • 额外的 "results" 配置存储所有运行的聚合结果,用于计算和显示聚合指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_TheBloke__Orca-2-13B-GPTQ_public", "harness_winogrande_5", split="train")

最新结果

以下是 2023-11-25T03:42:21.410226 运行的最新结果

python { "all": { "acc": 0.5887851314518572, "acc_stderr": 0.032958137391722146, "acc_norm": 0.5969185976587905, "acc_norm_stderr": 0.03368773395313244, "mc1": 0.38555691554467564, "mc1_stderr": 0.01703883901059167, "mc2": 0.5514098320774886, "mc2_stderr": 0.0160327733300155, "em": 0.42606963087248323, "em_stderr": 0.0050641847856105855, "f1": 0.5302139261744996, "f1_stderr": 0.004659796001509701 }, ... }

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2023_11_25T03_42_21.410226
      • path: **/details_harness|arc:challenge|25_2023-11-25T03-42-21.410226.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-11-25T03-42-21.410226.parquet
  • config_name: harness_drop_3

    • split: 2023_11_25T03_42_21.410226
      • path: **/details_harness|drop|3_2023-11-25T03-42-21.410226.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-11-25T03-42-21.410226.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_11_25T03_42_21.410226
      • path: **/details_harness|gsm8k|5_2023-11-25T03-42-21.410226.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-11-25T03-42-21.410226.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_11_25T03_42_21.410226
      • path: **/details_harness|hellaswag|10_2023-11-25T03-42-21.410226.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-11-25T03-42-21.410226.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_11_25T03_42_21.410226
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-11-25T03-42-21.410226.parquet
        • ...
    • split: latest
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-11-25T03-42-21.410226.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-11-25T03-42-21.410226.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