open-llm-leaderboard-old/details_Enno-Ai__ennodata-7b
收藏数据集概述
数据集摘要
该数据集是在对模型 Enno-Ai/ennodata-7b 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
数据集结构
- 数据集包含 64 个配置,每个配置对应一个评估任务。
- 数据集从 2 次运行中创建,每次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。
- 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示聚合指标在 Open LLM Leaderboard。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Enno-Ai__ennodata-7b", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-27T05:41:02.798297 运行的最新结果: python { "all": { "em": 0.0010486577181208054, "em_stderr": 0.0003314581465219155, "f1": 0.055922818791946494, "f1_stderr": 0.0012829613643597505, "acc": 0.3733497369811503, "acc_stderr": 0.008984164865569185 }, "harness|drop|3": { "em": 0.0010486577181208054, "em_stderr": 0.0003314581465219155, "f1": 0.055922818791946494, "f1_stderr": 0.0012829613643597505 }, "harness|gsm8k|5": { "acc": 0.037149355572403335, "acc_stderr": 0.005209516283073758 }, "harness|winogrande|5": { "acc": 0.7095501183898973, "acc_stderr": 0.01275881344806461 } }
配置详情
-
harness_arc_challenge_25
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-17T18:21:05.699051.parquet
-
harness_drop_3
- 分片: 2023_10_27T05_41_02.798297, latest
- 路径:
**/details_harness|drop|3_2023-10-27T05-41-02.798297.parquet
-
harness_gsm8k_5
- 分片: 2023_10_27T05_41_02.798297, latest
- 路径:
**/details_harness|gsm8k|5_2023-10-27T05-41-02.798297.parquet
-
harness_hellaswag_10
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hellaswag|10_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径: 包含多个子任务的详细路径
-
harness_hendrycksTest_abstract_algebra_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_anatomy_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_astronomy_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_business_ethics_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_college_biology_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_college_medicine_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_college_physics_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_computer_security_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_econometrics_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-17T18:21:05.699051.parquet
-
harness_hendrycksTest_elementary_mathematics_5
- 分片: 2023_08_17T18_21_05.699051, latest
- 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-17T18:21:05.699051.parquet
以上是数据集的详细概述,包括数据集的结构、加载示例、最新结果以及各个配置的详细信息。



