open-llm-leaderboard-old/details_gmonsoon__delta-4b-orange
收藏数据集概述
数据集摘要
该数据集是在评估模型 gmonsoon/delta-4b-orange 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。"train" 分割总是指向最新的结果。
数据集结构
数据集包含多个配置,每个配置对应不同的评估任务。以下是部分配置的详细信息:
配置示例
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2024_03_09T19_02_36.573006
- path:
- **/details_harness|arc:challenge|25_2024-03-09T19-02-36.573006.parquet
- path:
- split: latest
- path:
- **/details_harness|arc:challenge|25_2024-03-09T19-02-36.573006.parquet
- path:
- split: 2024_03_09T19_02_36.573006
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2024_03_09T19_02_36.573006
- path:
- **/details_harness|gsm8k|5_2024-03-09T19-02-36.573006.parquet
- path:
- split: latest
- path:
- **/details_harness|gsm8k|5_2024-03-09T19-02-36.573006.parquet
- path:
- split: 2024_03_09T19_02_36.573006
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2024_03_09T19_02_36.573006
- path:
- **/details_harness|hellaswag|10_2024-03-09T19-02-36.573006.parquet
- path:
- split: latest
- path:
- **/details_harness|hellaswag|10_2024-03-09T19-02-36.573006.parquet
- path:
- split: 2024_03_09T19_02_36.573006
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2024_03_09T19_02_36.573006
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-03-09T19-02-36.573006.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-03-09T19-02-36.573006.parquet
- path:
- split: 2024_03_09T19_02_36.573006
- data_files:
最新结果
以下是 2024-03-09T19:02:36.573006 运行 的最新结果:
python { "all": { "acc": 0.5672998500589107, "acc_stderr": 0.03374547957816994, "acc_norm": 0.568767763482637, "acc_norm_stderr": 0.0344408518430215, "mc1": 0.401468788249694, "mc1_stderr": 0.017160273901693654, "mc2": 0.5681886396364959, "mc2_stderr": 0.015914705006773194 }, "harness|arc:challenge|25": { "acc": 0.568259385665529, "acc_stderr": 0.014474591427196202, "acc_norm": 0.5887372013651877, "acc_norm_stderr": 0.014379441068522089 }, "harness|hellaswag|10": { "acc": 0.5881298546106354, "acc_stderr": 0.004911659884506146, "acc_norm": 0.7658832901812388, "acc_norm_stderr": 0.004225800787050875 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.25, "acc_stderr": 0.04351941398892446, "acc_norm": 0.25, "acc_norm_stderr": 0.04351941398892446 }, "harness|hendrycksTest-anatomy|5": { "acc": 0.4888888888888889, "acc_stderr": 0.04318275491977976, "acc_norm": 0.4888888888888889, "acc_norm_stderr": 0.04318275491977976 }, "harness|hendrycksTest-astronomy|5": { "acc": 0.5657894736842105, "acc_stderr": 0.040335656678483205, "acc_norm": 0.5657894736842105, "acc_norm_stderr": 0.040335656678483205 }, "harness|hendrycksTest-business_ethics|5": { "acc": 0.65, "acc_stderr": 0.047937248544110196, "acc_norm": 0.65, "acc_norm_stderr": 0.047937248544110196 }, "harness|hendrycksTest-clinical_knowledge|5": { "acc": 0.5811320754716981, "acc_stderr": 0.030365050829115208, "acc_norm": 0.5811320754716981, "acc_norm_stderr": 0.030365050829115208 }, "harness|hendrycksTest-college_biology|5": { "acc": 0.5833333333333334, "acc_stderr": 0.04122728707651282, "acc_norm": 0.5833333333333334, "acc_norm_stderr": 0.04122728707651282 }, "harness|hendrycksTest-college_chemistry|5": { "acc": 0.34, "acc_stderr": 0.04760952285695236, "acc_norm": 0.34, "acc_norm_stderr": 0.04760952285695236 }, "harness|hendrycksTest-college_computer_science|5": { "acc": 0.49, "acc_stderr": 0.05024183937956913, "acc_norm": 0.49, "acc_norm_stderr": 0.05024183937956913 }, "harness|hendrycksTest-college_mathematics|5": { "acc": 0.39, "acc_stderr": 0.04902071300001974, "acc_norm": 0.39, "acc_norm_stderr": 0.04902071300001974 }, "harness|hendrycksTest-college_medicine|5": { "acc": 0.5549132947976878, "acc_stderr": 0.03789401760283647, "acc_norm": 0.5549132947976878, "acc_norm_stderr": 0.03789401760283647 }, "harness|hendrycksTest-college_physics|5": { "acc": 0.29411764705882354, "acc_stderr": 0.04533838195929775, "acc_norm": 0.29411764705882354, "acc_norm_stderr": 0.04533838195929775 }, "harness|hendrycksTest-computer_security|5": { "acc": 0.7, "acc_stderr": 0.046056618647183814, "acc_norm": 0.7, "acc_norm_stderr": 0.046056618647183814 }, "harness|hendrycksTest-conceptual_physics|5": { "acc": 0.4808510638297872, "acc_stderr": 0.03266204299064678, "acc_norm": 0.4808510638297872, "acc_norm_stderr": 0.03266204299064678 }, "harness|hendrycksTest-econometrics|5": { "acc": 0.3333333333333333, "acc_stderr": 0.044346007015849245, "acc_norm": 0.3333333333333333, "acc_norm_stderr": 0.044346007015849245 }, "harness|hendrycksTest-electrical_engineering|5": { "acc": 0.41379310344827586, "acc_stderr": 0.04104269211806232, "acc_norm": 0.413793103



