five

open-llm-leaderboard-old/details_Azazelle__Sina-Thor-7b-Merge

收藏
Hugging Face2024-01-11 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Azazelle__Sina-Thor-7b-Merge
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Azazelle/Sina-Thor-7b-Merge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。数据集从1次运行中创建,每次运行可以在每个配置中找到,运行的时间戳作为分割名称。train分割始终指向最新的结果。此外,results配置存储了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。

该数据集是在Open LLM Leaderboard上对模型Azazelle/Sina-Thor-7b-Merge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。数据集从1次运行中创建,每次运行可以在每个配置中找到,运行的时间戳作为分割名称。train分割始终指向最新的结果。此外,results配置存储了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。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 Azazelle/Sina-Thor-7b-Merge 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集结构

  • 配置数量:63个配置,每个配置对应一个评估任务。
  • 运行次数:数据集来自1次运行。每个运行结果作为一个特定的分割存储在每个配置中,分割名称使用运行的时间戳。
  • 训练分割:"train" 分割始终指向最新的结果。
  • 结果配置: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

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Azazelle__Sina-Thor-7b-Merge", "harness_winogrande_5", split="train")

最新结果

以下是 2024-01-11T02:04:55.621660 运行 的最新结果:

python { "all": { "acc": 0.654717106968504, "acc_stderr": 0.03185313943141863, "acc_norm": 0.6553216710780829, "acc_norm_stderr": 0.032502309346968794, "mc1": 0.3353733170134639, "mc1_stderr": 0.01652753403966899, "mc2": 0.5000640698738921, "mc2_stderr": 0.015237006727015454 }, "harness|arc:challenge|25": { "acc": 0.6262798634812287, "acc_stderr": 0.014137708601759088, "acc_norm": 0.6621160409556314, "acc_norm_stderr": 0.01382204792228351 }, "harness|hellaswag|10": { "acc": 0.6745668193586934, "acc_stderr": 0.0046757891569776475, "acc_norm": 0.8569010157339175, "acc_norm_stderr": 0.003494581076398528 }, ... }

配置详情

  • 配置名称:harness_arc_challenge_25

    • 数据文件
      • 分割:2024_01_11T02_04_55.621660
        • 路径:**/details_harness|arc:challenge|25_2024-01-11T02-04-55.621660.parquet
      • 分割:latest
        • 路径:**/details_harness|arc:challenge|25_2024-01-11T02-04-55.621660.parquet
  • 配置名称:harness_gsm8k_5

    • 数据文件
      • 分割:2024_01_11T02_04_55.621660
        • 路径:**/details_harness|gsm8k|5_2024-01-11T02-04-55.621660.parquet
      • 分割:latest
        • 路径:**/details_harness|gsm8k|5_2024-01-11T02-04-55.621660.parquet
  • 配置名称:harness_hellaswag_10

    • 数据文件
      • 分割:2024_01_11T02_04_55.621660
        • 路径:**/details_harness|hellaswag|10_2024-01-11T02-04-55.621660.parquet
      • 分割:latest
        • 路径:**/details_harness|hellaswag|10_2024-01-11T02-04-55.621660.parquet
  • 配置名称:harness_hendrycksTest_5

    • 数据文件
      • 分割:2024_01_11T02_04_55.621660
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-management|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2024-01-11T02-04-55.621660.parquet
          • **/details_harness|hendrycksTest-professional_medicine|5_2024-01-11T02-04-55.621660.parquet
          • `**/details_
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