five

open-llm-leaderboard/details_Doctor-Shotgun__mythospice-70b

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_Doctor-Shotgun__mythospice-70b
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Doctor-Shotgun/mythospice-70b进行评估时自动创建的。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。

This dataset was automatically created during the evaluation of the model Doctor-Shotgun/mythospice-70b on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluation task. It is generated from two evaluation runs, with the results of each run serving as a dataset split under a specific configuration, where the split name uses the timestamp of the run. The 'train' split always points to the most recent results. Additionally, there is a configuration named 'results' that stores the aggregated results from all runs and is used to calculate and display the aggregated metrics on the Open LLM Leaderboard.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集来源

该数据集是在对模型 Doctor-Shotgun/mythospice-70b 进行评估运行期间自动创建的,评估结果显示在 Open LLM Leaderboard 上。

数据集结构

  • 配置数量:64个配置,每个配置对应一个评估任务。
  • 创建次数:数据集从2次运行中创建。每个运行结果作为一个特定的分割存储,分割名称使用运行的时间戳。
  • 最新结果:"train" 分割始终指向最新的结果。
  • 汇总结果:一个额外的配置 "results" 存储所有运行结果的汇总,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_Doctor-Shotgun__mythospice-70b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-24T21:51:42.689346 运行的最新结果: python { "all": { "em": 0.002726510067114094, "em_stderr": 0.0005340111700415905, "f1": 0.06940331375838925, "f1_stderr": 0.0014269735757716981, "acc": 0.5668306034144879, "acc_stderr": 0.011562556636019638 }, "harness|drop|3": { "em": 0.002726510067114094, "em_stderr": 0.0005340111700415905, "f1": 0.06940331375838925, "f1_stderr": 0.0014269735757716981 }, "harness|gsm8k|5": { "acc": 0.3009855951478393, "acc_stderr": 0.012634504465211199 }, "harness|winogrande|5": { "acc": 0.8326756116811366, "acc_stderr": 0.010490608806828079 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割:2023_10_10T17_34_08.268208
      • 路径:**/details_harness|arc:challenge|25_2023-10-10T17-34-08.268208.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-10-10T17-34-08.268208.parquet
  • harness_drop_3

    • 分割:2023_10_24T21_51_42.689346
      • 路径:**/details_harness|drop|3_2023-10-24T21-51-42.689346.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-24T21-51-42.689346.parquet
  • harness_gsm8k_5

    • 分割:2023_10_24T21_51_42.689346
      • 路径:**/details_harness|gsm8k|5_2023-10-24T21-51-42.689346.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-24T21-51-42.689346.parquet
  • harness_hellaswag_10

    • 分割:2023_10_10T17_34_08.268208
      • 路径:**/details_harness|hellaswag|10_2023-10-10T17-34-08.268208.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-10-10T17-34-08.268208.parquet
  • harness_hendrycksTest_5

    • 分割:2023_10_10T17_34_08.268208
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-management|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-10-10T17-34-08.268208.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-10-10T17-34-08.268208.parquet
        • `**/details_harness|hendrycksTest-professional_law|5_2023-10-10T17-34-08.268208.
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