open-llm-leaderboard-old/details_willyninja30__ARIA-70B-French
收藏数据集概述
数据集来源
该数据集是在评估模型 willyninja30/ARIA-70B-French 在 Open LLM Leaderboard 上的自动创建的。
数据集结构
- 数据集包含 64 个配置,每个配置对应一个评估任务。
- 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 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_willyninja30__ARIA-70B-French", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-25T03:07:36.932003 运行的最新结果: python { "all": { "em": 0.040373322147651006, "em_stderr": 0.0020157564185176837, "f1": 0.1050272651006715, "f1_stderr": 0.0023756238577676155, "acc": 0.5359600711595986, "acc_stderr": 0.011658939983913113 }, "harness|drop|3": { "em": 0.040373322147651006, "em_stderr": 0.0020157564185176837, "f1": 0.1050272651006715, "f1_stderr": 0.0023756238577676155 }, "harness|gsm8k|5": { "acc": 0.266868840030326, "acc_stderr": 0.012183780551887957 }, "harness|winogrande|5": { "acc": 0.8050513022888713, "acc_stderr": 0.011134099415938268 } }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_09_22T07_22_49.937285
- path: **/details_harness|arc:challenge|25_2023-09-22T07-22-49.937285.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-09-22T07-22-49.937285.parquet
- split: 2023_09_22T07_22_49.937285
-
config_name: harness_drop_3
- split: 2023_10_25T03_07_36.932003
- path: **/details_harness|drop|3_2023-10-25T03-07-36.932003.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-25T03-07-36.932003.parquet
- split: 2023_10_25T03_07_36.932003
-
config_name: harness_gsm8k_5
- split: 2023_10_25T03_07_36.932003
- path: **/details_harness|gsm8k|5_2023-10-25T03-07-36.932003.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-25T03-07-36.932003.parquet
- split: 2023_10_25T03_07_36.932003
-
config_name: harness_hellaswag_10
- split: 2023_09_22T07_22_49.937285
- path: **/details_harness|hellaswag|10_2023-09-22T07-22-49.937285.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-09-22T07-22-49.937285.parquet
- split: 2023_09_22T07_22_49.937285
-
config_name: harness_hendrycksTest_5
- split: 2023_09_22T07_22_49.937285
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-management|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-09-22T07-22-49.937285.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-09-22T07-22-49.937285.parquet
- **/
- path:
- split: 2023_09_22T07_22_49.937285



