open-llm-leaderboard/details_Corianas__gpt-j-6B-Dolly
收藏数据集概述
数据集来源
该数据集是在对模型 Corianas/gpt-j-6B-Dolly 进行评估运行期间自动创建的,评估结果展示在 Open LLM Leaderboard 上。
数据集结构
数据集由64个配置组成,每个配置对应一个评估任务。数据集从2次运行中创建,每次运行的结果可以在每个配置中找到,以运行的时间戳命名的特定分割形式存在。"train" 分割始终指向最新的结果。
数据集加载示例
以下是加载数据集详细信息的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Corianas__gpt-j-6B-Dolly", "harness_winogrande_5", split="train")
最新结果
以下是 2023-09-23T01:23:52.389948 运行 的最新结果: python { "all": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642965902, "f1": 0.08095008389261764, "f1_stderr": 0.0017837058432559263, "acc": 0.32814795356315596, "acc_stderr": 0.008022527306282314 }, "harness|drop|3": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642965902, "f1": 0.08095008389261764, "f1_stderr": 0.0017837058432559263 }, "harness|gsm8k|5": { "acc": 0.009097801364670205, "acc_stderr": 0.002615326510775673 }, "harness|winogrande|5": { "acc": 0.6471981057616417, "acc_stderr": 0.013429728101788956 } }
配置详情
以下是数据集的部分配置详情:
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2023_07_19T15_40_52.841362
- path: **/details_harness|arc:challenge|25_2023-07-19T15:40:52.841362.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-07-19T15:40:52.841362.parquet
- split: 2023_07_19T15_40_52.841362
- data_files:
-
config_name: harness_drop_3
- data_files:
- split: 2023_09_23T01_23_52.389948
- path: **/details_harness|drop|3_2023-09-23T01-23-52.389948.parquet
- split: latest
- path: **/details_harness|drop|3_2023-09-23T01-23-52.389948.parquet
- split: 2023_09_23T01_23_52.389948
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2023_09_23T01_23_52.389948
- path: **/details_harness|gsm8k|5_2023-09-23T01-23-52.389948.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-09-23T01-23-52.389948.parquet
- split: 2023_09_23T01_23_52.389948
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2023_07_19T15_40_52.841362
- path: **/details_harness|hellaswag|10_2023-07-19T15:40:52.841362.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-07-19T15:40:52.841362.parquet
- split: 2023_07_19T15_40_52.841362
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2023_07_19T15_40_52.841362
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-management|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T15:40:52.841362.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-07-19T15:40:52.
- path:
- split: 2023_07_19T15_40_52.841362
- data_files:



