five

open-llm-leaderboard/details_CHIH-HUNG__llama-2-13b-huangyt_Fintune_1_17w-q_k_v_o_proj

收藏
Hugging Face2023-10-18 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_CHIH-HUNG__llama-2-13b-huangyt_Fintune_1_17w-q_k_v_o_proj
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CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-q_k_v_o_proj时自动生成的,评估是在Open LLM Leaderboard上进行的。数据集包含64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,每次运行的结果作为特定的分割存储在配置中,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,还有一个名为results的配置,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
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 CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-q_k_v_o_projOpen LLM Leaderboard 上的运行过程中自动创建的。

数据集结构

  • 数据集包含 64 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_CHIH-HUNG__llama-2-13b-huangyt_Fintune_1_17w-q_k_v_o_proj", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-18T03:41:50.218488 运行的最新结果: python { "all": { "em": 0.35979446308724833, "em_stderr": 0.0049150348561349125, "f1": 0.39900062919463114, "f1_stderr": 0.004825664226319409, "acc": 0.46081554692422627, "acc_stderr": 0.01059065324860096 }, "harness|drop|3": { "em": 0.35979446308724833, "em_stderr": 0.0049150348561349125, "f1": 0.39900062919463114, "f1_stderr": 0.004825664226319409 }, "harness|gsm8k|5": { "acc": 0.14025777103866566, "acc_stderr": 0.009565108281428673 }, "harness|winogrande|5": { "acc": 0.7813733228097869, "acc_stderr": 0.011616198215773246 } }

配置详情

以下是数据集的配置详情:

  • config_name: harness_arc_challenge_25

    • split: 2023_09_04T02_19_44.261303
      • path: **/details_harness|arc:challenge|25_2023-09-04T02:19:44.261303.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-09-04T02:19:44.261303.parquet
  • config_name: harness_drop_3

    • split: 2023_10_18T03_41_50.218488
      • path: **/details_harness|drop|3_2023-10-18T03-41-50.218488.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-10-18T03-41-50.218488.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_10_18T03_41_50.218488
      • path: **/details_harness|gsm8k|5_2023-10-18T03-41-50.218488.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-10-18T03-41-50.218488.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_09_04T02_19_44.261303
      • path: **/details_harness|hellaswag|10_2023-09-04T02:19:44.261303.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-09-04T02:19:44.261303.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_09_04T02_19_44.261303
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-management|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-09-04T02:19:44.261303.parquet
        • **/details_harness|hendrycksTest-
二维码
社区交流群
二维码
科研交流群
商业服务