five

open-llm-leaderboard/details_WizardLM__WizardLM-13B-V1.2

收藏
Hugging Face2023-10-18 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_WizardLM__WizardLM-13B-V1.2
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WizardLM/WizardLM-13B-V1.2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,每次运行的结果存储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 WizardLM/WizardLM-13B-V1.2. It consists of 64 configurations, each corresponding to one evaluation task. The dataset is generated from two runs, with the results of each run stored as a split under its respective 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 across all runs, which is used to calculate and display aggregate metrics on the Open LLM Leaderboard.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

数据集结构

  • 配置数量: 64
  • 配置详情: 每个配置对应一个评估任务。
  • 数据来源: 从2次运行中创建。每个运行在每个配置中作为一个特定的分割存在,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: "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_WizardLM__WizardLM-13B-V1.2", "harness_winogrande_5", split="train")

最新结果

  • 来源: 最新结果来自 run 2023-10-18T23:07:01.737511
  • 指标: python { "all": { "em": 0.09133808724832215, "em_stderr": 0.002950304012601038, "f1": 0.1617292365771806, "f1_stderr": 0.0032231699829319426, "acc": 0.4269860152120696, "acc_stderr": 0.011021928189223498 }, "harness|drop|3": { "em": 0.09133808724832215, "em_stderr": 0.002950304012601038, "f1": 0.1617292365771806, "f1_stderr": 0.0032231699829319426 }, "harness|gsm8k|5": { "acc": 0.13495072024260804, "acc_stderr": 0.009411315282571171 }, "harness|winogrande|5": { "acc": 0.7190213101815311, "acc_stderr": 0.012632541095875825 } }

配置详情

  • harness_arc_challenge_25

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|arc:challenge|25_2023-07-26T14:20:40.943670.parquet
  • harness_drop_3

    • 分割: 2023_10_18T23_07_01.737511, latest
    • 路径: **/details_harness|drop|3_2023-10-18T23-07-01.737511.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_18T23_07_01.737511, latest
    • 路径: **/details_harness|gsm8k|5_2023-10-18T23-07-01.737511.parquet
  • harness_hellaswag_10

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hellaswag|10_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: 多个文件路径,例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-26T14:20:40.943670.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割: 2023_07_26T14_20_40.943670, latest
    • 路径: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-26T14:20:40.943670.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