open-llm-leaderboard/details_jukofyork__Eurus-70b-nca-fixed
收藏Hugging Face2024-04-16 更新2024-06-12 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_jukofyork__Eurus-70b-nca-fixed
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[jukofyork/Eurus-70b-nca-fixed](https://huggingface.co/jukofyork/Eurus-70b-nca-fixed)时自动生成的,评估是在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行的。数据集由63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,每次运行都可以在特定配置中找到,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,还有一个名为results的配置,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。README还提供了如何加载数据集中的运行细节的示例代码,并展示了最新的评估结果。
该数据集是在评估模型[jukofyork/Eurus-70b-nca-fixed](https://huggingface.co/jukofyork/Eurus-70b-nca-fixed)时自动生成的,评估是在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行的。数据集由63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,每次运行都可以在特定配置中找到,分割名称使用运行的时间戳。train分割始终指向最新的结果。此外,还有一个名为results的配置,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。README还提供了如何加载数据集中的运行细节的示例代码,并展示了最新的评估结果。
提供机构:
open-llm-leaderboard
原始信息汇总
数据集概述
数据集名称
- pretty_name: Evaluation run of jukofyork/Eurus-70b-nca-fixed
数据集描述
- dataset_summary: 该数据集是在评估模型jukofyork/Eurus-70b-nca-fixed运行期间自动创建的,用于Open LLM Leaderboard。
- 组成: 包含63个配置,每个配置对应一个评估任务。
- 创建来源: 数据集由1次运行创建,每次运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳命名。
- 额外配置: “results”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。
数据集加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_jukofyork__Eurus-70b-nca-fixed", "harness_winogrande_5", split="train")
最新结果
- 结果来源: latest results from run 2024-04-16T00:23:57.124428
- 结果内容: 包含多个任务的评估结果,如准确率(acc)、标准误差(acc_stderr)等。
数据集配置详情
配置列表
-
harness_arc_challenge_25
- 数据文件:
- split: 2024_04_16T00_23_57.124428
- path: /details_harness|arc:challenge|25_2024-04-16T00-23-57.124428.parquet
- split: latest
- path: /details_harness|arc:challenge|25_2024-04-16T00-23-57.124428.parquet
- split: 2024_04_16T00_23_57.124428
- 数据文件:
-
harness_gsm8k_5
- 数据文件:
- split: 2024_04_16T00_23_57.124428
- path: /details_harness|gsm8k|5_2024-04-16T00-23-57.124428.parquet
- split: latest
- path: /details_harness|gsm8k|5_2024-04-16T00-23-57.124428.parquet
- split: 2024_04_16T00_23_57.124428
- 数据文件:
-
harness_hellaswag_10
- 数据文件:
- split: 2024_04_16T00_23_57.124428
- path: /details_harness|hellaswag|10_2024-04-16T00-23-57.124428.parquet
- split: latest
- path: /details_harness|hellaswag|10_2024-04-16T00-23-57.124428.parquet
- split: 2024_04_16T00_23_57.124428
- 数据文件:
-
harness_hendrycksTest_5
- 数据文件:
- split: 2024_04_16T00_23_57.124428
- path:
- /details_harness|hendrycksTest-abstract_algebra|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-anatomy|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-astronomy|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-business_ethics|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-clinical_knowledge|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-college_biology|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-college_chemistry|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-college_computer_science|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-college_mathematics|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-college_medicine|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-college_physics|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-computer_security|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-conceptual_physics|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-econometrics|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-electrical_engineering|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-elementary_mathematics|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-formal_logic|5_2024-04-16T00-23-57.124428.parquet
- /details_harness|hendrycksTest-global_facts|5_2024-04-16T00-23-57.124428.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-04
- path:
- split: 2024_04_16T00_23_57.124428
- 数据文件:



