five

open-llm-leaderboard-old/details_deepseek-ai__deepseek-math-7b-instruct

收藏
Hugging Face2024-03-13 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_deepseek-ai__deepseek-math-7b-instruct
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deepseek-ai/deepseek-math-7b-instruct时自动生成的,包含63个配置,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在不同的split中,split以运行的时间戳命名。train split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,用于在Open LLM Leaderboard上计算和显示聚合指标。

该数据集是在评估模型deepseek-ai/deepseek-math-7b-instruct时自动生成的,包含63个配置,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在不同的split中,split以运行的时间戳命名。train split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,用于在Open LLM Leaderboard上计算和显示聚合指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集摘要

该数据集是在评估模型 deepseek-ai/deepseek-math-7b-instructOpen LLM Leaderboard 上的运行过程中自动创建的。数据集包含 63 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,每次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。

数据集结构

数据集包含多个配置,每个配置对应不同的评估任务。每个配置包含多个分割,包括特定时间戳的分割和最新的分割。

最新结果

以下是 最新结果 的摘要:

python { "all": { "acc": 0.554019303836701, "acc_stderr": 0.034497761386808885, "acc_norm": 0.5618809813979222, "acc_norm_stderr": 0.035247964085412954, "mc1": 0.29253365973072215, "mc1_stderr": 0.015925597445286165, "mc2": 0.40156731347428204, "mc2_stderr": 0.014934119039002425 }, "harness|arc:challenge|25": { "acc": 0.5008532423208191, "acc_stderr": 0.014611369529813269, "acc_norm": 0.5341296928327645, "acc_norm_stderr": 0.014577311315231108 }, "harness|hellaswag|10": { "acc": 0.5409281019717188, "acc_stderr": 0.004973036453863722, "acc_norm": 0.7149970125473013, "acc_norm_stderr": 0.004504932999736403 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.04793724854411021, "acc_norm": 0.35, "acc_norm_stderr": 0.04793724854411021 }, # 其他任务的结果... }

配置详情

数据集包含多个配置,每个配置对应不同的评估任务。以下是部分配置的示例:

  • config_name: harness_arc_challenge_25

    • 分割: 2024_03_12T07_42_48.023389
      • 路径: **/details_harness|arc:challenge|25_2024-03-12T07-42-48.023389.parquet
    • 分割: 2024_03_13T18_13_18.094811
      • 路径: **/details_harness|arc:challenge|25_2024-03-13T18-13-18.094811.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2024-03-13T18-13-18.094811.parquet
  • config_name: harness_gsm8k_5

    • 分割: 2024_03_12T07_42_48.023389
      • 路径: **/details_harness|gsm8k|5_2024-03-12T07-42-48.023389.parquet
    • 分割: 2024_03_13T18_13_18.094811
      • 路径: **/details_harness|gsm8k|5_2024-03-13T18-13-18.094811.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2024-03-13T18-13-18.094811.parquet
  • config_name: harness_hellaswag_10

    • 分割: 2024_03_12T07_42_48.023389
      • 路径: **/details_harness|hellaswag|10_2024-03-12T07-42-48.023389.parquet
    • 分割: 2024_03_13T18_13_18.094811
      • 路径: **/details_harness|hellaswag|10_2024-03-13T18-13-18.094811.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2024-03-13T18-13-18.094811.parquet
  • config_name: harness_hendrycksTest_5

    • 分割: 2024_03_12T07_42_48.023389
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-international_law|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-management|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-marketing|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2024-03-12T07-42-48.023389.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2024-03-12T07-42-48.023389.parquet
        • `**/details_harness|hendrycksTest-philosophy|5_2024-03-12T07-42-48.023389.
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