open-llm-leaderboard-old/details_AbacusResearch__jaLLAbi2-7b
收藏数据集概述
数据集简介
该数据集是在评估模型 AbacusResearch/jaLLAbi2-7b 在 Open LLM Leaderboard 上的运行过程中自动创建的。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1次运行创建,每个运行可以在每个配置中找到特定的拆分,拆分名称使用运行的时间戳。"train" 拆分始终指向最新的结果。
额外配置
一个额外的配置 "results" 存储了所有运行的聚合结果,用于计算和显示 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_AbacusResearch__jaLLAbi2-7b", "harness_winogrande_5", split="train")
最新结果
以下是 2024-02-20T08:55:37.448857 运行的最新结果:
python { "all": { "acc": 0.65503470601137, "acc_stderr": 0.03201407812269774, "acc_norm": 0.654286580962508, "acc_norm_stderr": 0.032683723254148166, "mc1": 0.5422276621787026, "mc1_stderr": 0.017440965712482125, "mc2": 0.701571568751529, "mc2_stderr": 0.014731078396613723 }, "harness|arc:challenge|25": { "acc": 0.6885665529010239, "acc_stderr": 0.01353247209985094, "acc_norm": 0.7167235494880546, "acc_norm_stderr": 0.013167478735134575 }, "harness|hellaswag|10": { "acc": 0.7123083051185023, "acc_stderr": 0.004517614647703243, "acc_norm": 0.8828918542123083, "acc_norm_stderr": 0.003208919510309925 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.0479372485441102, "acc_norm": 0.35, "acc_norm_stderr": 0.0479372485441102 }, ... }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2024_02_20T08_55_37.448857
- **/details_harness|arc:challenge|25_2024-02-20T08-55-37.448857.parquet
- split: latest
- **/details_harness|arc:challenge|25_2024-02-20T08-55-37.448857.parquet
- split: 2024_02_20T08_55_37.448857
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2024_02_20T08_55_37.448857
- **/details_harness|gsm8k|5_2024-02-20T08-55-37.448857.parquet
- split: latest
- **/details_harness|gsm8k|5_2024-02-20T08-55-37.448857.parquet
- split: 2024_02_20T08_55_37.448857
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2024_02_20T08_55_37.448857
- **/details_harness|hellaswag|10_2024-02-20T08-55-37.448857.parquet
- split: latest
- **/details_harness|hellaswag|10_2024-02-20T08-55-37.448857.parquet
- split: 2024_02_20T08_55_37.448857
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2024_02_20T08_55_37.448857
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-human_aging|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-international_law|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-management|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-marketing|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-nutrition|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-philosophy|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-prehistory|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2024-02-20T08-55-37.448857.parquet
- **/details_harness|hendrycksTest-professional_law|5_2024-02-20T08-55-37.44885
- split: 2024_02_20T08_55_37.448857
- data_files:



