open-llm-leaderboard-old/details_Josephgflowers__TinyLlama-Cinder-1.3B-Test.2
收藏数据集概述
数据集摘要
该数据集是在评估模型Josephgflowers/TinyLlama-Cinder-1.3B-Test.2在Open LLM Leaderboard上的自动创建的。数据集包含63个配置,每个配置对应一个评估任务。
数据集结构
- 配置数量:63个配置
- 数据来源:从1次运行中创建,每个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。
- 最新结果:"train"分割始终指向最新结果。
- 聚合结果: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_Josephgflowers__TinyLlama-Cinder-1.3B-Test.2", "harness_winogrande_5", split="train")
最新结果
这些是最新结果,来自2024-01-27T17:11:20.414206的运行: python { "all": { "acc": 0.2632532516237253, "acc_stderr": 0.03096899556861394, "acc_norm": 0.26383605058795256, "acc_norm_stderr": 0.03172701859516164, "mc1": 0.21909424724602203, "mc1_stderr": 0.014480038578757438, "mc2": 0.3798174614120003, "mc2_stderr": 0.01429160027055937 }, "harness|arc:challenge|25": { "acc": 0.31143344709897613, "acc_stderr": 0.013532472099850945, "acc_norm": 0.3370307167235495, "acc_norm_stderr": 0.01381347665290227 }, "harness|hellaswag|10": { "acc": 0.4422425811591316, "acc_stderr": 0.004956378590571539, "acc_norm": 0.5866361282613025, "acc_norm_stderr": 0.004914305798575694 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.27, "acc_stderr": 0.044619604333847415, "acc_norm": 0.27, "acc_norm_stderr": 0.044619604333847415 }, # 其他任务的结果... }
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分割:2024_01_27T17_11_20.414206
- 路径:
**/details_harness|arc:challenge|25_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:2024_01_27T17_11_20.414206
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分割:2024_01_27T17_11_20.414206
- 路径:
**/details_harness|gsm8k|5_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:2024_01_27T17_11_20.414206
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分割:2024_01_27T17_11_20.414206
- 路径:
**/details_harness|hellaswag|10_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:2024_01_27T17_11_20.414206
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分割:2024_01_27T17_11_20.414206
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-human_aging|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-international_law|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-management|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-marketing|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-nutrition|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-philosophy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-prehistory|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_law|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_psychology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-public_relations|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-security_studies|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-sociology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-virology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-world_religions|5_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-human_aging|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-international_law|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-management|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-marketing|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-nutrition|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-philosophy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-prehistory|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_law|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-professional_psychology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-public_relations|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-security_studies|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-sociology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-virology|5_2024-01-27T17-11-20.414206.parquet**/details_harness|hendrycksTest-world_religions|5_2024-01-27T17-11-20.414206.parquet
- 路径:
- 分割:2024_01_27T17_11_20.414206
- 数据文件:



