open-llm-leaderboard-old/details_NovoCode__Novocode7b-v3
收藏Hugging Face2024-01-26 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_NovoCode__Novocode7b-v3
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NovoCode/Novocode7b-v3进行评估时自动创建的。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分割始终指向最新结果。此外,results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上的聚合指标。
该数据集是在Open LLM Leaderboard上对模型NovoCode/Novocode7b-v3进行评估时自动创建的。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分割始终指向最新结果。此外,results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上的聚合指标。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集基本信息
- 名称: Evaluation run of NovoCode/Novocode7b-v3
- 来源: 该数据集是在模型 NovoCode/Novocode7b-v3 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
数据集组成
- 配置数量: 63个配置,每个配置对应一个评估任务。
- 创建次数: 从1次运行中创建。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。
- 最新结果: "train" 分割总是指向最新的结果。
- 结果汇总: 一个额外的配置 "results" 存储所有运行的汇总结果,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_NovoCode__Novocode7b-v3", "harness_winogrande_5", split="train")
最新结果
- 最新结果时间戳: 2024-01-26T12:46:59.252533
- 结果示例:
python
{
"all": {
"acc": 0.6159809063122001,
"acc_stderr": 0.03273859989879733,
"acc_norm": 0.6215791848788114,
"acc_norm_stderr": 0.03339933314489889,
"mc1": 0.32558139534883723,
"mc1_stderr": 0.016403989469907825,
"mc2": 0.4829352733126611,
"mc2_stderr": 0.016049866289528984
},
"harness|arc:challenge|25": {
"acc": 0.5546075085324232,
"acc_stderr": 0.014523987638344076,
"acc_norm": 0.5750853242320819,
"acc_norm_stderr": 0.014445698968520767
},
"harness|hellaswag|10": {
"acc": 0.6263692491535551,
"acc_stderr": 0.004827786289074841,
"acc_norm": 0.8116908982274448,
"acc_norm_stderr": 0.0039015979142464933
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.37,
"acc_stderr": 0.04852365870939099,
"acc_norm": 0.37,
"acc_norm_stderr": 0.04852365870939099
},
其他任务的结果...
}
配置详情
-
配置名称: harness_arc_challenge_25
- 数据文件:
- 分割: 2024_01_26T12_46_59.252533
- 路径: **/details_harness|arc:challenge|25_2024-01-26T12-46-59.252533.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-01-26T12-46-59.252533.parquet
- 分割: 2024_01_26T12_46_59.252533
- 数据文件:
-
配置名称: harness_gsm8k_5
- 数据文件:
- 分割: 2024_01_26T12_46_59.252533
- 路径: **/details_harness|gsm8k|5_2024-01-26T12-46-59.252533.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-01-26T12-46-59.252533.parquet
- 分割: 2024_01_26T12_46_59.252533
- 数据文件:
-
配置名称: harness_hellaswag_10
- 数据文件:
- 分割: 2024_01_26T12_46_59.252533
- 路径: **/details_harness|hellaswag|10_2024-01-26T12-46-59.252533.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-01-26T12-46-59.252533.parquet
- 分割: 2024_01_26T12_46_59.252533
- 数据文件:
-
配置名称: harness_hendrycksTest_5
- 数据文件:
- 分割: 2024_01_26T12_46_59.252533
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-01-26T12-46-59.252533.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024
- 路径:
- 分割: 2024_01_26T12_46_59.252533
- 数据文件:



