open-llm-leaderboard-old/details_AA051610__A0105
收藏Hugging Face2024-01-06 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_AA051610__A0105
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AA051610/A0105进行评估时自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。数据集包含1次运行的结果,每次运行在每个配置中表示为特定的拆分。train拆分始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。
该数据集是在Open LLM Leaderboard上对模型AA051610/A0105进行评估时自动创建的。它由63个配置组成,每个配置对应一个被评估的任务。数据集包含1次运行的结果,每次运行在每个配置中表示为特定的拆分。train拆分始终指向最新的结果。一个名为results的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示Open LLM Leaderboard上的聚合指标。可以使用`datasets`库中的`load_dataset`函数加载该数据集。
提供机构:
open-llm-leaderboard-old
原始信息汇总
数据集概述
数据集创建
- 创建背景:该数据集是在模型AA051610/A0105在Open LLM Leaderboard上的评估运行期间自动创建的。
- 数据集组成:包含63个配置,每个配置对应一个评估任务。
- 数据来源:数据集从1次运行中创建,每个运行在每个配置中作为一个特定的分片存在,分片名称使用运行的时间戳。"train"分片始终指向最新的结果。
数据集加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_AA051610__A0105", "harness_winogrande_5", split="train")
最新结果
- 结果来源:最新结果来自2024-01-06T02:57:13.678426的运行。
- 结果示例:
python
{
"all": {
"acc": 0.6820912387442206,
"acc_stderr": 0.031065747816855848,
"acc_norm": 0.6864043317313094,
"acc_norm_stderr": 0.031666952183494475,
"mc1": 0.3843329253365973,
"mc1_stderr": 0.0170287073012452,
"mc2": 0.5543558949846231,
"mc2_stderr": 0.016036294123592646
},
"harness|arc:challenge|25": {
"acc": 0.6117747440273038,
"acc_stderr": 0.014241614207414044,
"acc_norm": 0.621160409556314,
"acc_norm_stderr": 0.014175915490000326
},
"harness|hellaswag|10": {
"acc": 0.6291575383389763,
"acc_stderr": 0.004820431839600027,
"acc_norm": 0.8254331806413066,
"acc_norm_stderr": 0.003788203729346702
},
"harness|hendrycksTest-abstract_algebra|5": {
"acc": 0.34,
"acc_stderr": 0.04760952285695235,
"acc_norm": 0.34,
"acc_norm_stderr": 0.04760952285695235
},
更多任务结果...
}
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分片:2024_01_06T02_57_13.678426
- 路径:
**/details_harness|arc:challenge|25_2024-01-06T02-57-13.678426.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|arc:challenge|25_2024-01-06T02-57-13.678426.parquet
- 路径:
- 分片:2024_01_06T02_57_13.678426
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分片:2024_01_06T02_57_13.678426
- 路径:
**/details_harness|gsm8k|5_2024-01-06T02-57-13.678426.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|gsm8k|5_2024-01-06T02-57-13.678426.parquet
- 路径:
- 分片:2024_01_06T02_57_13.678426
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分片:2024_01_06T02_57_13.678426
- 路径:
**/details_harness|hellaswag|10_2024-01-06T02-57-13.678426.parquet
- 路径:
- 分片:latest
- 路径:
**/details_harness|hellaswag|10_2024-01-06T02-57-13.678426.parquet
- 路径:
- 分片:2024_01_06T02_57_13.678426
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分片:2024_01_06T02_57_13.678426
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-human_aging|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-international_law|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-management|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-marketing|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-nutrition|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-philosophy|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-prehistory|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-professional_law|5_2024-01-06T02-57-13.678426.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-01-06T02-57-13.678426.parquet- `**/details_harness|hendrycksTest-professional_psychology|5_
- 路径:
- 分片:2024_01_06T02_57_13.678426
- 数据文件:



