open-llm-leaderboard-old/details_gmonsoon__OpenMia-Indo-Mistral-7b-v3-refined
收藏数据集概述
数据集简介
该数据集是在评估模型 gmonsoon/OpenMia-Indo-Mistral-7b-v3-refined 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
数据集由 2 次运行创建,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。
额外配置
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_gmonsoon__OpenMia-Indo-Mistral-7b-v3-refined", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的摘要:
python { "all": { "acc": 0.6291464441848997, "acc_stderr": 0.0325052901080867, "acc_norm": 0.6303599490256852, "acc_norm_stderr": 0.03317036353110983, "mc1": 0.3769889840881273, "mc1_stderr": 0.01696551757893035, "mc2": 0.5394835979252358, "mc2_stderr": 0.015320518165942699 }, "harness|arc:challenge|25": { "acc": 0.6049488054607508, "acc_stderr": 0.014285898292938167, "acc_norm": 0.64419795221843, "acc_norm_stderr": 0.01399057113791876 }, "harness|hellaswag|10": { "acc": 0.640211113324039, "acc_stderr": 0.004789575163418651, "acc_norm": 0.842162915753834, "acc_norm_stderr": 0.0036384306206139337 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.3, "acc_stderr": 0.046056618647183814, "acc_norm": 0.3, "acc_norm_stderr": 0.046056618647183814 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- 分割: 2024_02_04T17_47_56.805589
- 路径: **/details_harness|arc:challenge|25_2024-02-04T17-47-56.805589.parquet
- 分割: 2024_02_04T19_41_28.521258
- 路径: **/details_harness|arc:challenge|25_2024-02-04T19-41-28.521258.parquet
- 分割: latest
- 路径: **/details_harness|arc:challenge|25_2024-02-04T19-41-28.521258.parquet
- 分割: 2024_02_04T17_47_56.805589
-
config_name: harness_gsm8k_5
- 分割: 2024_02_04T17_47_56.805589
- 路径: **/details_harness|gsm8k|5_2024-02-04T17-47-56.805589.parquet
- 分割: 2024_02_04T19_41_28.521258
- 路径: **/details_harness|gsm8k|5_2024-02-04T19-41-28.521258.parquet
- 分割: latest
- 路径: **/details_harness|gsm8k|5_2024-02-04T19-41-28.521258.parquet
- 分割: 2024_02_04T17_47_56.805589
-
config_name: harness_hellaswag_10
- 分割: 2024_02_04T17_47_56.805589
- 路径: **/details_harness|hellaswag|10_2024-02-04T17-47-56.805589.parquet
- 分割: 2024_02_04T19_41_28.521258
- 路径: **/details_harness|hellaswag|10_2024-02-04T19-41-28.521258.parquet
- 分割: latest
- 路径: **/details_harness|hellaswag|10_2024-02-04T19-41-28.521258.parquet
- 分割: 2024_02_04T17_47_56.805589
-
config_name: harness_hendrycksTest_5
- 分割: 2024_02_04T17_47_56.805589
- 路径:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-management|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-02-04T17-47-56.805589.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-04T17-47-56.805589.
- 路径:
- 分割: 2024_02_04T17_47_56.805589



