five

open-llm-leaderboard/details_Locutusque__gpt2-conversational-or-qa

收藏
Hugging Face2023-09-17 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_Locutusque__gpt2-conversational-or-qa
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型Locutusque/gpt2-conversational-or-qa在Open LLM Leaderboard上的评估运行期间自动创建的。它由64个配置组成,每个配置对应一个被评估的任务。数据集是从2次运行中生成的,每次运行在每个配置中表示为特定的分割。train分割始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用Python中的datasets库加载该数据集。

This dataset was automatically created during the evaluation run of the model Locutusque/gpt2-conversational-or-qa on the Open LLM Leaderboard. It consists of 64 configurations, each corresponding to one evaluated task. The dataset is generated from two separate runs, where each run is represented as a specific split under each configuration. The `train` split always points to the most up-to-date results. An additional configuration named `results` stores the aggregated results across all runs, which are used to calculate and display the aggregate metrics on the Open LLM Leaderboard. This dataset can be loaded using the `datasets` library in Python.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

该数据集是在评估模型 Locutusque/gpt2-conversational-or-qaOpen LLM Leaderboard 上的自动创建的。数据集包含 64 个配置,每个配置对应一个评估任务。

数据集结构

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。

数据集加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Locutusque__gpt2-conversational-or-qa", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果来自 run 2023-09-17T06:39:40.166876 的摘要:

python { "all": { "em": 0.00041946308724832214, "em_stderr": 0.00020969854707829385, "f1": 0.015460360738255055, "f1_stderr": 0.0006333702020804492, "acc": 0.25610125343097334, "acc_stderr": 0.007403477156790923 }, "harness|drop|3": { "em": 0.00041946308724832214, "em_stderr": 0.00020969854707829385, "f1": 0.015460360738255055, "f1_stderr": 0.0006333702020804492 }, "harness|gsm8k|5": { "acc": 0.000758150113722517, "acc_stderr": 0.0007581501137225174 }, "harness|winogrande|5": { "acc": 0.5114443567482242, "acc_stderr": 0.014048804199859329 } }

配置详情

数据集包含多个配置,每个配置对应不同的任务和运行时间戳。以下是部分配置的示例:

  • config_name: harness_arc_challenge_25

    • split: 2023_07_18T16_08_01.149355
      • path: **/details_harness|arc:challenge|25_2023-07-18T16:08:01.149355.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-07-18T16:08:01.149355.parquet
  • config_name: harness_drop_3

    • split: 2023_09_17T06_39_40.166876
      • path: **/details_harness|drop|3_2023-09-17T06-39-40.166876.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-09-17T06-39-40.166876.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_09_17T06_39_40.166876
      • path: **/details_harness|gsm8k|5_2023-09-17T06-39-40.166876.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-09-17T06-39-40.166876.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_07_18T16_08_01.149355
      • path: **/details_harness|hellaswag|10_2023-07-18T16:08:01.149355.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-07-18T16:08:01.149355.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_07_18T16_08_01.149355
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-management|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-07-18T16:08:01.149355.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_202
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