five

open-llm-leaderboard/details_migtissera__Synthia-34B-v1.2

收藏
Hugging Face2023-09-18 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_migtissera__Synthia-34B-v1.2
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migtissera/Synthia-34B-v1.2在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。数据集由一次运行生成,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。一个名为results的额外配置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。README还提供了如何使用Hugging Face datasets库加载运行细节的示例。
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 migtissera/Synthia-34B-v1.2Open LLM Leaderboard 上的自动创建的。数据集包含 61 个配置,每个配置对应一个评估任务。

数据集结构

数据集由 1 次运行创建,每个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。

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 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_migtissera__Synthia-34B-v1.2", "harness_truthfulqa_mc_0", split="train")

最新结果

这些是最新结果,来自 2023-09-18T20:05:34.645170 的运行。注意,可能有其他任务的结果在仓库中,如果连续的评估没有覆盖相同的任务。你可以在 "results" 和每个评估的 "latest" 分割中找到每个任务的结果。

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2023_09_18T20_05_34.645170
      • path: [**/details_harness|arc:challenge|25_2023-09-18T20-05-34.645170.parquet]
    • split: latest
      • path: [**/details_harness|arc:challenge|25_2023-09-18T20-05-34.645170.parquet]
  • config_name: harness_hellaswag_10

    • split: 2023_09_18T20_05_34.645170
      • path: [**/details_harness|hellaswag|10_2023-09-18T20-05-34.645170.parquet]
    • split: latest
      • path: [**/details_harness|hellaswag|10_2023-09-18T20-05-34.645170.parquet]
  • config_name: harness_hendrycksTest_5

    • split: 2023_09_18T20_05_34.645170
      • path: [**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-anatomy|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-astronomy|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-business_ethics|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-college_biology|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-college_chemistry|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-college_computer_science|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-college_mathematics|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-college_medicine|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-college_physics|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-computer_security|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-econometrics|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-formal_logic|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-global_facts|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-high_school_biology|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-high_school_geography|5_2023-09-18T20-05-34.645170.parquet, **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-18T20-05-34.645170.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