five

open-llm-leaderboard-old/details_jeiku__Garbage_9B

收藏
Hugging Face2024-04-03 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_jeiku__Garbage_9B
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jeiku/Garbage_9B进行评估时自动创建的。它由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中生成的,每次运行在每个配置中表示为特定的拆分,使用运行的时间戳命名。"train"拆分始终指向最新结果。一个额外的配置"results"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。可以使用`datasets`库中的`load_dataset`函数加载该数据集。

该数据集是在Open LLM Leaderboard上对模型jeiku/Garbage_9B进行评估时自动创建的。它由63个配置组成,每个配置对应一个评估任务。数据集是从1次运行中生成的,每次运行在每个配置中表示为特定的拆分,使用运行的时间戳命名。"train"拆分始终指向最新结果。一个额外的配置"results"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。可以使用`datasets`库中的`load_dataset`函数加载该数据集。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集信息

  • 数据集名称: Evaluation run of jeiku/Garbage_9B
  • 数据集来源: 自动创建于模型 jeiku/Garbage_9BOpen LLM Leaderboard 的评估运行期间。
  • 数据集组成: 包含 63 个配置,每个配置对应一个评估任务。
  • 数据集创建: 从 1 次运行中创建,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。"train" 分割始终指向最新结果。
  • 额外配置: "results" 配置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。

最新结果

  • 最新结果时间戳: 2024-04-03T01:02:39.315171
  • 结果示例: python { "all": { "acc": 0.6417247545580523, "acc_stderr": 0.032174976054508406, "acc_norm": 0.6448916698706535, "acc_norm_stderr": 0.0328204246636684, "mc1": 0.3182374541003672, "mc1_stderr": 0.016305988648920605, "mc2": 0.4786566619579289, "mc2_stderr": 0.014555828379762867 }, "harness|arc:challenge|25": { "acc": 0.5895904436860068, "acc_stderr": 0.014374922192642664, "acc_norm": 0.636518771331058, "acc_norm_stderr": 0.014056207319068283 }, "harness|hellaswag|10": { "acc": 0.6276638119896435, "acc_stderr": 0.004824393076826621, "acc_norm": 0.8297151961760606, "acc_norm_stderr": 0.0037511461503847653 }, # 更多任务结果... }

配置信息

  • 配置名称: harness_arc_challenge_25

    • 数据文件:
      • 分割: 2024_04_03T01_02_39.315171
        • 路径: **/details_harness|arc:challenge|25_2024-04-03T01-02-39.315171.parquet
      • 分割: latest
        • 路径: **/details_harness|arc:challenge|25_2024-04-03T01-02-39.315171.parquet
  • 配置名称: harness_gsm8k_5

    • 数据文件:
      • 分割: 2024_04_03T01_02_39.315171
        • 路径: **/details_harness|gsm8k|5_2024-04-03T01-02-39.315171.parquet
      • 分割: latest
        • 路径: **/details_harness|gsm8k|5_2024-04-03T01-02-39.315171.parquet
  • 配置名称: harness_hellaswag_10

    • 数据文件:
      • 分割: 2024_04_03T01_02_39.315171
        • 路径: **/details_harness|hellaswag|10_2024-04-03T01-02-39.315171.parquet
      • 分割: latest
        • 路径: **/details_harness|hellaswag|10_2024-04-03T01-02-39.315171.parquet
  • 配置名称: harness_hendrycksTest_5

    • 数据文件:
      • 分割: 2024_04_03T01_02_39.315171
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-management|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-professional_medicine|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-professional_psychology|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-public_relations|5_2024-04-03T01-02-39.315171.parquet
          • **/details_harness|hendrycksTest-security_studies|5_2024-04-03T01-02-3
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