open-llm-leaderboard/details_Faradaylab__Aria-70B
收藏Hugging Face2023-08-27 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_Faradaylab__Aria-70B
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Faradaylab/Aria-70B进行评估时自动创建的。数据集由61个配置组成,每个配置对应一个评估任务。数据集包含1次运行的结果,每次运行在每种配置中表示为特定的分割,train分割始终指向最新结果。此外,名为results的配置存储了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。README还提供了如何使用`datasets`库中的`load_dataset`函数加载数据集的示例,并包含了特定运行的最新结果。
提供机构:
open-llm-leaderboard
原始信息汇总
数据集概述
数据集组成
- 该数据集包含61个配置,每个配置对应一个评估任务。
- 数据集从1次运行中创建,每个运行的结果可以在每个配置中找到,以运行的时间戳命名的特定分割形式存储。
- "train"分割始终指向最新的结果。
- 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_Faradaylab__Aria-70B", "harness_truthfulqa_mc_0", split="train")
最新结果
- 最新结果来自2023-08-26T09:05:40.294272的运行,包含多个任务的评估结果。
配置详情
-
harness_arc_challenge_25
- 数据文件路径:
**/details_harness|arc:challenge|25_2023-08-26T09:05:40.294272.parquet
- 数据文件路径:
-
harness_hellaswag_10
- 数据文件路径:
**/details_harness|hellaswag|10_2023-08-26T09:05:40.294272.parquet
- 数据文件路径:
-
harness_hendrycksTest_5
- 数据文件路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-anatomy|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-astronomy|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-college_biology|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-college_physics|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-computer_security|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-econometrics|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-global_facts|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-08-26T09:05:40.294272.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-26T09:05:40.294272.parquet- `**/details_harness|hendrycksTest-high_school_macro
- 数据文件路径:



