open-llm-leaderboard-old/details_BreadAi__PM_modelV2
收藏数据集概述
数据集是在评估模型 BreadAi/PM_modelV2 在 Open LLM Leaderboard 上的运行过程中自动创建的。
数据集组成
- 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" 分割总是指向最新的结果。
- 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。
加载数据集示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_BreadAi__PM_modelV2", "harness_winogrande_5", split="train")
最新结果
以下是 2023-09-16T22:18:34.076880 运行的最新结果: python { "all": { "em": 0.0009437919463087249, "em_stderr": 0.00031446531194131066, "f1": 0.004450503355704698, "f1_stderr": 0.00041742876892271523, "acc": 0.2478295185477506, "acc_stderr": 0.007025978032038445 }, "harness|drop|3": { "em": 0.0009437919463087249, "em_stderr": 0.00031446531194131066, "f1": 0.004450503355704698, "f1_stderr": 0.00041742876892271523 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|winogrande|5": { "acc": 0.4956590370955012, "acc_stderr": 0.01405195606407689 } }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_07_19T19_16_43.765981
- path: **/details_harness|arc:challenge|25_2023-07-19T19:16:43.765981.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-07-19T19:16:43.765981.parquet
- split: 2023_07_19T19_16_43.765981
-
config_name: harness_drop_3
- split: 2023_09_16T22_18_34.076880
- path: **/details_harness|drop|3_2023-09-16T22-18-34.076880.parquet
- split: latest
- path: **/details_harness|drop|3_2023-09-16T22-18-34.076880.parquet
- split: 2023_09_16T22_18_34.076880
-
config_name: harness_gsm8k_5
- split: 2023_09_16T22_18_34.076880
- path: **/details_harness|gsm8k|5_2023-09-16T22-18-34.076880.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-09-16T22-18-34.076880.parquet
- split: 2023_09_16T22_18_34.076880
-
config_name: harness_hellaswag_10
- split: 2023_07_19T19_16_43.765981
- path: **/details_harness|hellaswag|10_2023-07-19T19:16:43.765981.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-07-19T19:16:43.765981.parquet
- split: 2023_07_19T19_16_43.765981
-
config_name: harness_hendrycksTest_5
- split: 2023_07_19T19_16_43.765981
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-management|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycksTest-professional_medicine|5_2023-07-19T19:16:43.765981.parquet
- **/details_harness|hendrycks
- path:
- split: 2023_07_19T19_16_43.765981



