five

open-llm-leaderboard-old/details_DreadPoor__Satyr-7B-Model_Stock

收藏
Hugging Face2024-04-06 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_DreadPoor__Satyr-7B-Model_Stock
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DreadPoor/Satyr-7B-Model_Stock进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都可以在特定配置的特定分割中找到,分割以运行的时间戳命名。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,并用于计算和显示Open LLM Leaderboard上的聚合指标。

该数据集是在Open LLM Leaderboard上对模型DreadPoor/Satyr-7B-Model_Stock进行评估时自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都可以在特定配置的特定分割中找到,分割以运行的时间戳命名。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

数据集简介

该数据集是在评估模型DreadPoor/Satyr-7B-Model_StockOpen LLM Leaderboard上的运行过程中自动创建的。

数据集组成

  • 数据集包含63个配置,每个配置对应一个评估任务。
  • 数据集由1次运行创建,每次运行的详细结果存储在特定的分割中,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_DreadPoor__Satyr-7B-Model_Stock", "harness_winogrande_5", split="train")

最新结果

以下是2024-04-06T03:21:27.554118运行的最新结果: python { "all": { "acc": 0.6538228091662829, "acc_stderr": 0.03204349764412167, "acc_norm": 0.6545608990736203, "acc_norm_stderr": 0.032693748107169206, "mc1": 0.4663402692778458, "mc1_stderr": 0.017463793867168106, "mc2": 0.6376075140129857, "mc2_stderr": 0.015484912688455579 }, "harness|arc:challenge|25": { "acc": 0.6612627986348123, "acc_stderr": 0.01383056892797433, "acc_norm": 0.6860068259385665, "acc_norm_stderr": 0.013562691224726302 }, "harness|hellaswag|10": { "acc": 0.6976697868950408, "acc_stderr": 0.004583289072937751, "acc_norm": 0.8696474805815575, "acc_norm_stderr": 0.003360027661765394 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.34, "acc_stderr": 0.047609522856952365, "acc_norm": 0.34, "acc_norm_stderr": 0.047609522856952365 }, ... }

配置详情

  • harness_arc_challenge_25

    • 分割: 2024_04_06T03_21_27.554118
    • 路径: **/details_harness|arc:challenge|25_2024-04-06T03-21-27.554118.parquet
    • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2024-04-06T03-21-27.554118.parquet
  • harness_gsm8k_5

    • 分割: 2024_04_06T03_21_27.554118
    • 路径: **/details_harness|gsm8k|5_2024-04-06T03-21-27.554118.parquet
    • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2024-04-06T03-21-27.554118.parquet
  • harness_hellaswag_10

    • 分割: 2024_04_06T03_21_27.554118
    • 路径: **/details_harness|hellaswag|10_2024-04-06T03-21-27.554118.parquet
    • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2024-04-06T03-21-27.554118.parquet
  • harness_hendrycksTest_5

    • 分割: 2024_04_06T03_21_27.554118
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2024-04-06T03-21-27.554118.parquet ...
    • 分割: latest
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2024-04-06T03-21-27.554118.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2024-04-06T03-21-27.554118.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