five

open-llm-leaderboard/details_TheBloke__WizardLM-13B-V1-1-SuperHOT-8K-GPTQ

收藏
Hugging Face2023-10-28 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_TheBloke__WizardLM-13B-V1-1-SuperHOT-8K-GPTQ
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TheBloke/WizardLM-13B-V1-1-SuperHOT-8K-GPTQ在Open LLM Leaderboard上的评估运行期间自动创建的。数据集由64个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行在每个配置中表示为特定的分割。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
原始信息汇总

数据集概述

该数据集是在评估模型 TheBloke/WizardLM-13B-V1-1-SuperHOT-8K-GPTQOpen LLM Leaderboard 上的自动创建的。

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 2 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。
  • 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

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_TheBloke__WizardLM-13B-V1-1-SuperHOT-8K-GPTQ", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-28T21:00:02.304492 运行的最新结果

python { "all": { "em": 0.22158137583892618, "em_stderr": 0.004253171428083824, "f1": 0.28616296140939684, "f1_stderr": 0.004276937020149761, "acc": 0.3751559533333772, "acc_stderr": 0.007270592555507228 }, "harness|drop|3": { "em": 0.22158137583892618, "em_stderr": 0.004253171428083824, "f1": 0.28616296140939684, "f1_stderr": 0.004276937020149761 }, "harness|gsm8k|5": { "acc": 0.006823351023502654, "acc_stderr": 0.0022675371022544783 }, "harness|winogrande|5": { "acc": 0.7434885556432518, "acc_stderr": 0.012273648008759979 } }

配置详情

  • harness_arc_challenge_25

    • 分割: 2023_09_11T17_32_08.880546
    • 路径: **/details_harness|arc:challenge|25_2023-09-11T17-32-08.880546.parquet
    • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2023-09-11T17-32-08.880546.parquet
  • harness_drop_3

    • 分割: 2023_10_28T21_00_02.304492
    • 路径: **/details_harness|drop|3_2023-10-28T21-00-02.304492.parquet
    • 分割: latest
    • 路径: **/details_harness|drop|3_2023-10-28T21-00-02.304492.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_28T21_00_02.304492
    • 路径: **/details_harness|gsm8k|5_2023-10-28T21-00-02.304492.parquet
    • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2023-10-28T21-00-02.304492.parquet
  • harness_hellaswag_10

    • 分割: 2023_09_11T17_32_08.880546
    • 路径: **/details_harness|hellaswag|10_2023-09-11T17-32-08.880546.parquet
    • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2023-09-11T17-32-08.880546.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_09_11T17_32_08.880546
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-management|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-11T17-32-08.880546.parquet
      • **/details_harness|hendrycksTest-professional_law|5_2023-0
二维码
社区交流群
二维码
科研交流群
商业服务