open-llm-leaderboard/details_LTC-AI-Labs__Guanaco-Vicuna-7B-L2
收藏数据集概述
数据集来源
该数据集是在对模型 LTC-AI-Labs/Guanaco-Vicuna-7B-L2 进行评估运行期间自动创建的,评估结果发布在 Open LLM Leaderboard 上。
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割存在,分割名称使用运行的时间戳。
- "train" 分割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_LTC-AI-Labs__Guanaco-Vicuna-7B-L2", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-29T06:29:02.987953 运行的最新结果: python { "all": { "em": 0.0014681208053691276, "em_stderr": 0.00039210421902985767, "f1": 0.056828859060402796, "f1_stderr": 0.0013179206618636607, "acc": 0.416677387679193, "acc_stderr": 0.0097821448230569 }, "harness|drop|3": { "em": 0.0014681208053691276, "em_stderr": 0.00039210421902985767, "f1": 0.056828859060402796, "f1_stderr": 0.0013179206618636607 }, "harness|gsm8k|5": { "acc": 0.07960576194086429, "acc_stderr": 0.007455924338676278 }, "harness|winogrande|5": { "acc": 0.7537490134175217, "acc_stderr": 0.012108365307437523 } }
配置详情
以下是数据集的配置详情:
-
harness_arc_challenge_25
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|arc:challenge|25_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_drop_3
- 分割:2023_10_29T06_29_02.987953
- 路径:
**/details_harness|drop|3_2023-10-29T06-29-02.987953.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|drop|3_2023-10-29T06-29-02.987953.parquet
- 路径:
- 分割:2023_10_29T06_29_02.987953
-
harness_gsm8k_5
- 分割:2023_10_29T06_29_02.987953
- 路径:
**/details_harness|gsm8k|5_2023-10-29T06-29-02.987953.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-29T06-29-02.987953.parquet
- 路径:
- 分割:2023_10_29T06_29_02.987953
-
harness_hellaswag_10
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hellaswag|10_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-anatomy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-astronomy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_biology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_physics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-computer_security|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-econometrics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-global_facts|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-human_aging|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-international_law|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-management|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-marketing|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-nutrition|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-philosophy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-prehistory|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_accounting|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_law|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_medicine|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_psychology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-public_relations|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-security_studies|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-sociology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-virology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-world_religions|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-anatomy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-astronomy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_biology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-college_physics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-computer_security|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-econometrics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-global_facts|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-human_aging|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-international_law|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-management|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-marketing|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-nutrition|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-philosophy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-prehistory|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_accounting|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_law|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_medicine|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-professional_psychology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-public_relations|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-security_studies|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-sociology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-virology|5_2023-10-01T13-18-10.170951.parquet**/details_harness|hendrycksTest-world_religions|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_anatomy_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_astronomy_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_college_biology_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_college_physics_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_computer_security_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_econometrics_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
-
harness_hendrycksTest_elementary_mathematics_5
- 分割:2023_10_01T13_18_10.170951
- 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-01T13-18-10.170951.parquet
- 路径:
- 分割:2023_10_01T13_18_10.170951
以上是数据集的详细配置信息,包括各个配置的名称、分割和路径。



