five

open-llm-leaderboard/details_zarakiquemparte__zarafusionex-1.2-l2-7b

收藏
Hugging Face2023-10-23 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_zarakiquemparte__zarafusionex-1.2-l2-7b
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型zarakiquemparte/zarafusionex-1.2-l2-7b时自动生成的,主要用于Open LLM Leaderboard的评估任务。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,每次运行的结果存储在不同的split中,split以时间戳命名。此外,数据集还包含一个名为"results"的配置,用于存储所有运行的聚合结果,并在Open LLM Leaderboard上显示聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集来源

该数据集是在评估模型 zarakiquemparte/zarafusionex-1.2-l2-7bOpen LLM Leaderboard 上的自动创建的。

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 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_zarakiquemparte__zarafusionex-1.2-l2-7b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-23T14:21:00.502633 运行 的最新结果: python { "all": { "em": 0.13590604026845637, "em_stderr": 0.0035094536867653074, "f1": 0.22546770134228172, "f1_stderr": 0.003699492065854956, "acc": 0.4165529242035385, "acc_stderr": 0.009960183652638432 }, "harness|drop|3": { "em": 0.13590604026845637, "em_stderr": 0.0035094536867653074, "f1": 0.22546770134228172, "f1_stderr": 0.003699492065854956 }, "harness|gsm8k|5": { "acc": 0.08567096285064443, "acc_stderr": 0.007709218855882758 }, "harness|winogrande|5": { "acc": 0.7474348855564326, "acc_stderr": 0.012211148449394107 } }

配置详情

以下是数据集的部分配置详情:

  • config_name: harness_arc_challenge_25

    • data_files:
      • split: 2023_09_22T00_24_36.284847
        • path: **/details_harness|arc:challenge|25_2023-09-22T00-24-36.284847.parquet
      • split: latest
        • path: **/details_harness|arc:challenge|25_2023-09-22T00-24-36.284847.parquet
  • config_name: harness_drop_3

    • data_files:
      • split: 2023_10_23T14_21_00.502633
        • path: **/details_harness|drop|3_2023-10-23T14-21-00.502633.parquet
      • split: latest
        • path: **/details_harness|drop|3_2023-10-23T14-21-00.502633.parquet
  • config_name: harness_gsm8k_5

    • data_files:
      • split: 2023_10_23T14_21_00.502633
        • path: **/details_harness|gsm8k|5_2023-10-23T14-21-00.502633.parquet
      • split: latest
        • path: **/details_harness|gsm8k|5_2023-10-23T14-21-00.502633.parquet
  • config_name: harness_hellaswag_10

    • data_files:
      • split: 2023_09_22T00_24_36.284847
        • path: **/details_harness|hellaswag|10_2023-09-22T00-24-36.284847.parquet
      • split: latest
        • path: **/details_harness|hellaswag|10_2023-09-22T00-24-36.284847.parquet
  • config_name: harness_hendrycksTest_5

    • data_files:
      • split: 2023_09_22T00_24_36.284847
        • path:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-international_law|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-management|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-marketing|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2023-09-22T00-24-36.284847.parquet
          • **/details_harness|hendrycksTest-
二维码
社区交流群
二维码
科研交流群
商业服务