open-llm-leaderboard-old/details_WizardLM__WizardCoder-15B-V1.0
收藏数据集概述
数据集名称
Evaluation run of WizardLM/WizardCoder-15B-V1.0
数据集描述
该数据集是在评估模型 WizardLM/WizardCoder-15B-V1.0 在 Open LLM Leaderboard 上的运行过程中自动创建的。
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。每次运行可以在每个配置中找到特定的分割,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 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_WizardLM__WizardCoder-15B-V1.0", "harness_winogrande_5", split="train")
最新结果
这些是最新结果(来自2023-10-18T05:39:03.080415的运行): python { "all": { "em": 0.10171979865771812, "em_stderr": 0.003095624755865799, "f1": 0.1654624580536908, "f1_stderr": 0.0033020160713134682, "acc": 0.2864625625234491, "acc_stderr": 0.008973810218487487 }, "harness|drop|3": { "em": 0.10171979865771812, "em_stderr": 0.003095624755865799, "f1": 0.1654624580536908, "f1_stderr": 0.0033020160713134682 }, "harness|gsm8k|5": { "acc": 0.02122820318423048, "acc_stderr": 0.003970449129848635 }, "harness|winogrande|5": { "acc": 0.5516969218626677, "acc_stderr": 0.013977171307126338 } }
配置详情
以下是数据集的配置详情:
配置列表
harness_arc_challenge_25harness_drop_3harness_gsm8k_5harness_hellaswag_10harness_hendrycksTest_5harness_hendrycksTest_abstract_algebra_5harness_hendrycksTest_anatomy_5harness_hendrycksTest_astronomy_5harness_hendrycksTest_business_ethics_5harness_hendrycksTest_clinical_knowledge_5harness_hendrycksTest_college_biology_5harness_hendrycksTest_college_chemistry_5harness_hendrycksTest_college_computer_science_5harness_hendrycksTest_college_mathematics_5harness_hendrycksTest_college_medicine_5harness_hendrycksTest_college_physics_5harness_hendrycksTest_computer_security_5harness_hendrycksTest_conceptual_physics_5harness_hendrycksTest_econometrics_5harness_hendrycksTest_electrical_engineering_5harness_hendrycksTest_elementary_mathematics_5harness_hendrycksTest_formal_logic_5harness_hendrycksTest_global_facts_5harness_hendrycksTest_high_school_biology_5harness_hendrycksTest_high_school_chemistry_5harness_hendrycksTest_high_school_computer_science_5harness_hendrycksTest_high_school_european_history_5harness_hendrycksTest_high_school_geography_5harness_hendrycksTest_high_school_government_and_politics_5harness_hendrycksTest_high_school_macroeconomics_5harness_hendrycksTest_high_school_mathematics_5harness_hendrycksTest_high_school_microeconomics_5harness_hendrycksTest_high_school_physics_5harness_hendrycksTest_high_school_psychology_5harness_hendrycksTest_high_school_statistics_5harness_hendrycksTest_high_school_us_history_5harness_hendrycksTest_high_school_world_history_5harness_hendrycksTest_human_aging_5harness_hendrycksTest_human_sexuality_5harness_hendrycksTest_international_law_5harness_hendrycksTest_jurisprudence_5harness_hendrycksTest_logical_fallacies_5harness_hendrycksTest_machine_learning_5harness_hendrycksTest_management_5harness_hendrycksTest_marketing_5harness_hendrycksTest_medical_genetics_5harness_hendrycksTest_miscellaneous_5harness_hendrycksTest_moral_disputes_5harness_hendrycksTest_moral_scenarios_5harness_hendrycksTest_nutrition_5harness_hendrycksTest_philosophy_5harness_hendrycksTest_prehistory_5harness_hendrycksTest_professional_accounting_5harness_hendrycksTest_professional_law_5harness_hendrycksTest_professional_medicine_5harness_hendrycksTest_professional_psychology_5harness_hendrycksTest_public_relations_5harness_hendrycksTest_security_studies_5harness_hendrycksTest_sociology_5harness_hendrycksTest_us_foreign_policy_5harness_hendrycksTest_virology_5harness_hendrycksTest_world_religions_5
配置详情
每个配置包含多个分割,每个分割对应不同的运行时间戳和路径。例如:
-
harness_arc_challenge_25- 分割
2023_07_19T20_24_20.327625- 路径:
**/details_harness|arc:challenge|25_2023-07-19T20:24:20.327625.parquet
- 路径:
- 分割
latest- 路径:
**/details_harness|arc:challenge|25_2023-07-19T20:24:20.327625.parquet
- 路径:
- 分割
-
harness_drop_3- 分割
2023_10_18T05_39_03.080415- 路径:
**/details_harness|drop|3_2023-10-18T05-39-03.080415.parquet
- 路径:
- 分割
latest- 路径:
**/details_harness|drop|3_2023-10-18T05-39-03.080415.parquet
- 路径:
- 分割
-
harness_gsm8k_5- 分割
2023_10_18T05_39_03.080415- 路径:
**/details_harness|gsm8k|5_2023-10-18T05-39-03.080415.parquet
- 路径:
- 分割
latest- 路径:
**/details_harness|gsm8k|5_2023-10-18T05-39-03.080415.parquet
- 路径:
- 分割
-
harness_hellaswag_10- 分割
2023_07_19T20_24_20.327625- 路径:
**/details_harness|hellaswag|10_2023-07-19T20:24:20.327625.parquet
- 路径:
- 分割
latest- 路径:
**/details_harness|hellaswag|10_2023-07-19T20:24:20.327625.parquet
- 路径:
- 分割
-
harness_hendrycksTest_5- 分割
2023_07_19T20_24_20.327625- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-formal_logic|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-global_facts|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T20:24:20.327625.parquet - 路径:
**/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T20:24:20.327625.parquet - 路径:`**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T20:2
- 路径:
- 分割



