open-llm-leaderboard-old/details_prithivida__Asimov-7B-v1
收藏Hugging Face2023-11-19 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_prithivida__Asimov-7B-v1
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prithivida/Asimov-7B-v1进行评估运行时自动创建的。数据集由64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分割始终指向最新结果。此外,还有一个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。README文件还包含一个Python代码片段,用于加载数据集,并提供了特定运行的最新结果。
该数据集是在Open LLM Leaderboard上对模型prithivida/Asimov-7B-v1进行评估运行时自动创建的。数据集由64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分割始终指向最新结果。此外,还有一个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。README文件还包含一个Python代码片段,用于加载数据集,并提供了特定运行的最新结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集名称
- 名称: Evaluation run of prithivida/Asimov-7B-v1
数据集组成
- 配置数量: 64个配置
- 每个配置对应: 一个评估任务
- 创建来源: 从1次运行中创建
- 分割命名: 使用运行的时间戳命名
- 训练分割: 总是指向最新结果
- 额外配置: "results"配置存储所有聚合结果,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_prithivida__Asimov-7B-v1_public", "harness_winogrande_5", split="train")
最新结果
- 结果来源: 最新结果来自2023-11-19T10:40:19.617701的运行
- 结果示例:
python
{
"all": {
"acc": 0.5591327615228756,
"acc_stderr": 0.033793433613618404,
"acc_norm": 0.5679554479286705,
"acc_norm_stderr": 0.034576211690701054,
"mc1": 0.3463892288861689,
"mc1_stderr": 0.01665699710912514,
"mc2": 0.5114755425083032,
"mc2_stderr": 0.015500857240755488,
"em": 0.004928691275167785,
"em_stderr": 0.0007171872517059772,
"f1": 0.06691170302013415,
"f1_stderr": 0.0015363127511980274
},
"harness|arc:challenge|25": {
"acc": 0.5460750853242321,
"acc_stderr": 0.014549221105171864,
"acc_norm": 0.590443686006826,
"acc_norm_stderr": 0.01437035863247244
},
"harness|hellaswag|10": {
"acc": 0.6097390957976498,
"acc_stderr": 0.004868117598481945,
"acc_norm": 0.8004381597291377,
"acc_norm_stderr": 0.00398854190214743
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.35,
"acc_stderr": 0.047937248544110196,
"acc_norm": 0.35,
"acc_norm_stderr": 0.047937248544110196
},
其他任务结果省略...
}
配置详情
-
配置名称: harness_arc_challenge_25
- 分割: 2023_11_19T10_40_19.617701
- 路径: **/details_harness|arc:challenge|25_2023-11-19T10-40-19.617701.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2023-11-19T10-40-19.617701.parquet
- 分割: 2023_11_19T10_40_19.617701
-
配置名称: harness_drop_3
- 分割: 2023_11_19T10_40_19.617701
- 路径: **/details_harness|drop|3_2023-11-19T10-40-19.617701.parquet
- 分割: latest
- 路径: **/details_harness|drop|3_2023-11-19T10-40-19.617701.parquet
- 分割: 2023_11_19T10_40_19.617701
-
配置名称: harness_gsm8k_5
- 分割: 2023_11_19T10_40_19.617701
- 路径: **/details_harness|gsm8k|5_2023-11-19T10-40-19.617701.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2023-11-19T10-40-19.617701.parquet
- 分割: 2023_11_19T10_40_19.617701
-
配置名称: harness_hellaswag_10
- 分割: 2023_11_19T10_40_19.617701
- 路径: **/details_harness|hellaswag|10_2023-11-19T10-40-19.617701.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2023-11-19T10-40-19.617701.parquet
- 分割: 2023_11_19T10_40_19.617701
-
配置名称: harness_hendrycksTest_5
- 分割: 2023_11_19T10_40_19.617701
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-management|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-11-19T10-40-19.617701.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-11-19T10-40-19.617701.parquet
- **/details_har
- 路径:
- 分割: 2023_11_19T10_40_19.617701



