open-llm-leaderboard-old/details_OpenBuddy__openbuddy-mistral2-7b-v20.1-32k
收藏数据集概述
数据集来源
该数据集是在对模型 OpenBuddy/openbuddy-mistral2-7b-v20.1-32k 进行评估运行时自动创建的,评估结果发布在 Open LLM Leaderboard 上。
数据集组成
数据集包含 63 个配置,每个配置对应一个评估任务。数据集由 1 次运行创建,每个运行在每个配置中都有一个特定的分割,分割名称使用运行的时间戳。"train" 分割始终指向最新的结果。
结果汇总
一个额外的配置 "results" 存储了所有运行的汇总结果,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_OpenBuddy__openbuddy-mistral2-7b-v20.1-32k", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的摘要:
python { "all": { "acc": 0.596817703139589, "acc_stderr": 0.03306526234430382, "acc_norm": 0.599574556453419, "acc_norm_stderr": 0.0337329949500657, "mc1": 0.36474908200734396, "mc1_stderr": 0.01685096106172011, "mc2": 0.5296972473987958, "mc2_stderr": 0.015203500502245027 }, "harness|arc:challenge|25": { "acc": 0.5068259385665529, "acc_stderr": 0.014610029151379813, "acc_norm": 0.5349829351535836, "acc_norm_stderr": 0.01457558392201967 }, "harness|hellaswag|10": { "acc": 0.5850428201553476, "acc_stderr": 0.004917076726623795, "acc_norm": 0.7776339374626569, "acc_norm_stderr": 0.00414985930060491 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.33, "acc_stderr": 0.04725815626252605, "acc_norm": 0.33, "acc_norm_stderr": 0.04725815626252605 }, # 其他任务的结果... }
配置详情
数据集包含多个配置,每个配置对应不同的评估任务和数据文件路径。以下是部分配置示例:
-
harness_arc_challenge_25:
- 数据文件路径:
**/details_harness|arc:challenge|25_2024-03-27T19-50-35.792936.parquet
- 数据文件路径:
-
harness_gsm8k_5:
- 数据文件路径:
**/details_harness|gsm8k|5_2024-03-27T19-50-35.792936.parquet
- 数据文件路径:
-
harness_hellaswag_10:
- 数据文件路径:
**/details_harness|hellaswag|10_2024-03-27T19-50-35.792936.parquet
- 数据文件路径:
-
harness_hendrycksTest_5:
- 数据文件路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-anatomy|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-astronomy|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-college_biology|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-college_physics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-computer_security|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-econometrics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-global_facts|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-human_aging|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-international_law|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-management|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-marketing|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-nutrition|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-philosophy|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-prehistory|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-professional_accounting|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-professional_law|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-professional_medicine|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-professional_psychology|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-public_relations|5_2024-03-27T19-50-35.792936.parquet**/details_harness|hendrycksTest-security_studies|5_2024-03-27T19-50-35.792936.parquet- `**/details_harness|hendrycksTest-sociology|5_2024-03-27T19-50-35.
- 数据文件路径:



