five

open-llm-leaderboard-old/details_radm__Philosophy-Platypus2-13b

收藏
Hugging Face2023-08-29 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_radm__Philosophy-Platypus2-13b
下载链接
链接失效反馈
官方服务:
资源简介:
数据集是在模型[radm/Philosophy-Platypus2-13b](https://huggingface.co/radm/Philosophy-Platypus2-13b)在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上的评估运行期间自动创建的。数据集由61个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都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"存储了所有运行的聚合结果,并用于计算和显示[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上的聚合指标。

数据集是在模型[radm/Philosophy-Platypus2-13b](https://huggingface.co/radm/Philosophy-Platypus2-13b)在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上的评估运行期间自动创建的。数据集由61个配置组成,每个配置对应一个评估任务。数据集是从1次运行中创建的,每次运行都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"存储了所有运行的聚合结果,并用于计算和显示[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/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

数据集信息

  • 名称: Evaluation run of radm/Philosophy-Platypus2-13b
  • 创建目的: 该数据集是在评估模型 radm/Philosophy-Platypus2-13bOpen LLM Leaderboard 上的运行过程中自动创建的。
  • 组成: 数据集包含61个配置,每个配置对应一个评估任务。
  • 创建次数: 数据集从1次运行中创建。每个运行可以在每个配置中找到一个特定的分割,分割名称使用运行的时间戳。"train" 分割总是指向最新的结果。
  • 额外配置: 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示在 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_radm__Philosophy-Platypus2-13b", "harness_truthfulqa_mc_0", split="train")

最新结果

  • 最新结果时间戳: 2023-08-29T00:45:24.163346
  • 聚合结果: python { "all": { "acc": 0.5437981691869808, "acc_stderr": 0.03484311795554624, "acc_norm": 0.547878610439407, "acc_norm_stderr": 0.034826606717822575, "mc1": 0.24357405140758873, "mc1_stderr": 0.015026354824910782, "mc2": 0.37335488461829447, "mc2_stderr": 0.014112790281285795 }, "harness|arc:challenge|25": { "acc": 0.5477815699658704, "acc_stderr": 0.014544519880633822, "acc_norm": 0.5861774744027304, "acc_norm_stderr": 0.014392730009221004 }, "harness|hellaswag|10": { "acc": 0.5828520215096594, "acc_stderr": 0.004920800313232742, "acc_norm": 0.785202150965943, "acc_norm_stderr": 0.0040984271589492634 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.34, "acc_stderr": 0.04760952285695235, "acc_norm": 0.34, "acc_norm_stderr": 0.04760952285695235 }, # 其他任务的结果省略... }

配置详情

  • 配置名称: harness_arc_challenge_25

    • 数据文件:
      • 分割: 2023_08_29T00_45_24.163346
        • 路径: **/details_harness|arc:challenge|25_2023-08-29T00:45:24.163346.parquet
      • 分割: latest
        • 路径: **/details_harness|arc:challenge|25_2023-08-29T00:45:24.163346.parquet
  • 配置名称: harness_hellaswag_10

    • 数据文件:
      • 分割: 2023_08_29T00_45_24.163346
        • 路径: **/details_harness|hellaswag|10_2023-08-29T00:45:24.163346.parquet
      • 分割: latest
        • 路径: **/details_harness|hellaswag|10_2023-08-29T00:45:24.163346.parquet
  • 配置名称: harness_hendrycksTest

    • 数据文件:
      • 分割: 2023_08_29T00_45_24.163346
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-international_law|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-management|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-marketing|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-professional_medicine|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-professional_psychology|5_2023-08-29T00:45:24.163346.parquet
          • **/details_harness|hendrycksTest-public_relations|5_2023-08-29T00:45:24.163346.parquet
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