open-llm-leaderboard-old/details_boomerchan__magpie-13b
收藏数据集概述
数据集简介
该数据集是在评估模型 boomerchan/magpie-13b 在 Open LLM Leaderboard 上的自动创建的。
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,用于计算和显示聚合指标在 Open LLM Leaderboard 上。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_boomerchan__magpie-13b", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-27T04:34:42.967550 运行的最新结果: python { "all": { "em": 0.14272231543624161, "em_stderr": 0.003582171317651424, "f1": 0.20778418624161069, "f1_stderr": 0.0036307604368272656, "acc": 0.4548027044477143, "acc_stderr": 0.01080662148135179 }, "harness|drop|3": { "em": 0.14272231543624161, "em_stderr": 0.003582171317651424, "f1": 0.20778418624161069, "f1_stderr": 0.0036307604368272656 }, "harness|gsm8k|5": { "acc": 0.14480667172100076, "acc_stderr": 0.009693234799052706 }, "harness|winogrande|5": { "acc": 0.7647987371744278, "acc_stderr": 0.011920008163650877 } }
配置详情
以下是部分配置及其数据文件路径:
-
config_name: harness_arc_challenge_25
- split: 2023_10_03T11_48_49.581129
- path: **/details_harness|arc:challenge|25_2023-10-03T11-48-49.581129.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-10-03T11-48-49.581129.parquet
- split: 2023_10_03T11_48_49.581129
-
config_name: harness_drop_3
- split: 2023_10_27T04_34_42.967550
- path: **/details_harness|drop|3_2023-10-27T04-34-42.967550.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-27T04-34-42.967550.parquet
- split: 2023_10_27T04_34_42.967550
-
config_name: harness_gsm8k_5
- split: 2023_10_27T04_34_42.967550
- path: **/details_harness|gsm8k|5_2023-10-27T04-34-42.967550.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-27T04-34-42.967550.parquet
- split: 2023_10_27T04_34_42.967550
-
config_name: harness_hellaswag_10
- split: 2023_10_03T11_48_49.581129
- path: **/details_harness|hellaswag|10_2023-10-03T11-48-49.581129.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-10-03T11-48-49.581129.parquet
- split: 2023_10_03T11_48_49.581129
-
config_name: harness_hendrycksTest_5
- split: 2023_10_03T11_48_49.581129
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-management|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-10-03T11-48-49.581129.parquet
- **/details_harness|hendrycksTest-professional_medicine|
- path:
- split: 2023_10_03T11_48_49.581129



