five

open-llm-leaderboard-old/details_nicholasKluge__Aira-Instruct-774M

收藏
Hugging Face2023-08-27 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_nicholasKluge__Aira-Instruct-774M
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nicholasKluge/Aira-Instruct-774M期间自动创建的,包含61个配置,每个配置对应一个评估任务。数据集由2次运行创建,每次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命名。此外,还有一个名为results的配置,存储了运行的所有聚合结果,用于计算和显示Open LLM Leaderboard上的聚合指标。

该数据集是在评估模型nicholasKluge/Aira-Instruct-774M期间自动创建的,包含61个配置,每个配置对应一个评估任务。数据集由2次运行创建,每次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命名。此外,还有一个名为results的配置,存储了运行的所有聚合结果,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 nicholasKluge/Aira-Instruct-774M 进行评估时自动创建的,用于 Open LLM Leaderboard

数据集结构

  • 配置数量:61个配置,每个配置对应一个评估任务。
  • 运行次数:数据集由2次运行创建。每次运行在每个配置中都有一个特定的分割,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,用于计算和显示聚合指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_nicholasKluge__Aira-Instruct-774M", "harness_truthfulqa_mc_0", split="train")

最新结果

以下是 2023-08-10T09:21:24.434701 运行的最新结果

python { "all": { "acc": 0.2509346011019478, "acc_stderr": 0.031496346014330656, "acc_norm": 0.25261641064354023, "acc_norm_stderr": 0.03150876208925782, "mc1": 0.2521419828641371, "mc1_stderr": 0.015201522246299972, "mc2": 0.42741407501534695, "mc2_stderr": 0.01548514736552057 }, "harness|arc:challenge|25": { "acc": 0.24488054607508533, "acc_stderr": 0.012566273985131358, "acc_norm": 0.2815699658703072, "acc_norm_stderr": 0.013143376735009024 }, "harness|hellaswag|10": { "acc": 0.3481378211511651, "acc_stderr": 0.004754063867700179, "acc_norm": 0.4106751643098984, "acc_norm_stderr": 0.004909509538525157 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.26, "acc_stderr": 0.04408440022768081, "acc_norm": 0.26, "acc_norm_stderr": 0.04408440022768081 }, # 其他任务的结果... }

配置详情

  • config_name: harness_arc_challenge_25

    • splits:
      • 2023_08_10T09_20_56.838686:
        • **/details_harness|arc:challenge|25_2023-08-10T09:20:56.838686.parquet
      • 2023_08_10T09_21_24.434701:
        • **/details_harness|arc:challenge|25_2023-08-10T09:21:24.434701.parquet
      • latest:
        • **/details_harness|arc:challenge|25_2023-08-10T09:21:24.434701.parquet
  • config_name: harness_hellaswag_10

    • splits:
      • 2023_08_10T09_20_56.838686:
        • **/details_harness|hellaswag|10_2023-08-10T09:20:56.838686.parquet
      • 2023_08_10T09_21_24.434701:
        • **/details_harness|hellaswag|10_2023-08-10T09:21:24.434701.parquet
      • latest:
        • **/details_harness|hellaswag|10_2023-08-10T09:21:24.434701.parquet
  • config_name: harness_hendrycksTest_5

    • splits:
      • 2023_08_10T09_20_56.838686:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-10T09:20:56.838686.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-08-10T09:20:56.838686
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