open-llm-leaderboard/details_dvruette__llama-13b-pretrained
收藏数据集概述
数据集简介
该数据集是在评估模型 dvruette/llama-13b-pretrained 在 Open LLM Leaderboard 上的运行过程中自动创建的。数据集包含64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" 存储所有运行结果的聚合信息,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_dvruette__llama-13b-pretrained", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的示例: python { "all": { "em": 0.19431627516778524, "em_stderr": 0.004052066229872751, "f1": 0.25224412751677777, "f1_stderr": 0.004066214952392991, "acc": 0.46513107858970915, "acc_stderr": 0.01097629037543693 }, "harness|drop|3": { "em": 0.19431627516778524, "em_stderr": 0.004052066229872751, "f1": 0.25224412751677777, "f1_stderr": 0.004066214952392991 }, "harness|gsm8k|5": { "acc": 0.1607278241091736, "acc_stderr": 0.010116708586037183 }, "harness|winogrande|5": { "acc": 0.7695343330702447, "acc_stderr": 0.011835872164836676 } }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2023_07_19T18_55_00.882635
- path: **/details_harness|arc:challenge|25_2023-07-19T18:55:00.882635.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-07-19T18:55:00.882635.parquet
- split: 2023_07_19T18_55_00.882635
- data_files:
-
config_name: harness_drop_3
- data_files:
- split: 2023_10_18T17_33_50.415201
- path: **/details_harness|drop|3_2023-10-18T17-33-50.415201.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-18T17-33-50.415201.parquet
- split: 2023_10_18T17_33_50.415201
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2023_10_18T17_33_50.415201
- path: **/details_harness|gsm8k|5_2023-10-18T17-33-50.415201.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-18T17-33-50.415201.parquet
- split: 2023_10_18T17_33_50.415201
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2023_07_19T18_55_00.882635
- path: **/details_harness|hellaswag|10_2023-07-19T18:55:00.882635.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-07-19T18:55:00.882635.parquet
- split: 2023_07_19T18_55_00.882635
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2023_07_19T18_55_00.882635
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-management|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-philosophy|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-prehistory|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-professional_law|5_2023-07-19T18:55:00.882635.parquet
- **/details_harness|hendrycksTest-professional_
- path:
- split: 2023_07_19T18_55_00.882635
- data_files:



