five

open-llm-leaderboard-old/details_openchat__openchat_v3.1

收藏
Hugging Face2023-10-16 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_openchat__openchat_v3.1
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openchat/openchat_v3.1时自动创建的,包含64个配置,每个配置对应一个评估任务。数据集由3次运行生成,每次运行的结果存储在不同的分割中,且最新的结果存储在train分割中。此外,数据集还包含一个名为results的配置,用于存储所有运行的聚合结果。

该数据集是在评估模型openchat/openchat_v3.1时自动创建的,包含64个配置,每个配置对应一个评估任务。数据集由3次运行生成,每次运行的结果存储在不同的分割中,且最新的结果存储在train分割中。此外,数据集还包含一个名为results的配置,用于存储所有运行的聚合结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集名称

Evaluation run of openchat/openchat_v3.1

数据集描述

该数据集是在openchat/openchat_v3.1模型在Open LLM Leaderboard上的评估运行期间自动创建的。

数据集组成

数据集由64个配置组成,每个配置对应一个评估任务。数据集从3次运行中创建,每次运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。"train"分片始终指向最新的结果。

额外配置

一个额外的配置"results"存储了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_openchat__openchat_v3.1", "harness_winogrande_5", split="train")

最新结果

以下是2023-10-16T02:39:54.553691运行的最新结果: python { "all": { "em": 0.0016778523489932886, "em_stderr": 0.00041913301788269345, "f1": 0.06259228187919454, "f1_stderr": 0.001365935795409535, "acc": 0.45020712996200873, "acc_stderr": 0.010730538116775 }, "harness|drop|3": { "em": 0.0016778523489932886, "em_stderr": 0.00041913301788269345, "f1": 0.06259228187919454, "f1_stderr": 0.001365935795409535 }, "harness|gsm8k|5": { "acc": 0.1379833206974981, "acc_stderr": 0.009499777327746841 }, "harness|winogrande|5": { "acc": 0.7624309392265194, "acc_stderr": 0.011961298905803162 } }

配置详情

  • config_name: harness_arc_challenge_25

    • data_files:
      • split: 2023_08_02T17_45_13.943818
        • path: **/details_harness|arc:challenge|25_2023-08-02T17:45:13.943818.parquet
      • split: latest
        • path: **/details_harness|arc:challenge|25_2023-08-02T17:45:13.943818.parquet
  • config_name: harness_drop_3

    • data_files:
      • split: 2023_09_24T04_16_26.631092
        • path: **/details_harness|drop|3_2023-09-24T04-16-26.631092.parquet
      • split: 2023_10_16T02_39_54.553691
        • path: **/details_harness|drop|3_2023-10-16T02-39-54.553691.parquet
      • split: latest
        • path: **/details_harness|drop|3_2023-10-16T02-39-54.553691.parquet
  • config_name: harness_gsm8k_5

    • data_files:
      • split: 2023_09_24T04_16_26.631092
        • path: **/details_harness|gsm8k|5_2023-09-24T04-16-26.631092.parquet
      • split: 2023_10_16T02_39_54.553691
        • path: **/details_harness|gsm8k|5_2023-10-16T02-39-54.553691.parquet
      • split: latest
        • path: **/details_harness|gsm8k|5_2023-10-16T02-39-54.553691.parquet
  • config_name: harness_hellaswag_10

    • data_files:
      • split: 2023_08_02T17_45_13.943818
        • path: **/details_harness|hellaswag|10_2023-08-02T17:45:13.943818.parquet
      • split: latest
        • path: **/details_harness|hellaswag|10_2023-08-02T17:45:13.943818.parquet
  • config_name: harness_hendrycksTest_5

    • data_files:
      • split: 2023_08_02T17_45_13.943818
        • path:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-international_law|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-management|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-marketing|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-02T17:45:13.943818.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2023-08-02T17:45:13.943818
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