open-llm-leaderboard/details_project-baize__baize-v2-13b
收藏Hugging Face2023-10-12 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_project-baize__baize-v2-13b
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在模型project-baize/baize-v2-13b的评估运行中自动创建的,用于Open LLM Leaderboard的评估。数据集由64个配置组成,每个配置对应一个评估任务。数据集是从2次运行中创建的,每次运行的结果都作为一个特定的分割存储在配置中,分割以运行的时间戳命名。train分割始终指向最新的结果。此外,还有一个名为results的配置,用于存储所有运行的聚合结果,并用于计算和显示在Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总
数据集概述
该数据集是在对模型 project-baize/baize-v2-13b 进行评估运行期间自动创建的。数据集包含64个配置,每个配置对应一个评估任务。数据集从2次运行中创建,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。
数据集结构
数据集包含以下配置:
-
harness_arc_challenge_25
- 分割:
2023_07_18T16_42_29.519016 - 路径:
**/details_harness|arc:challenge|25_2023-07-18T16:42:29.519016.parquet - 分割:
latest - 路径:
**/details_harness|arc:challenge|25_2023-07-18T16:42:29.519016.parquet
- 分割:
-
harness_drop_3
- 分割:
2023_10_12T16_13_42.802156 - 路径:
**/details_harness|drop|3_2023-10-12T16-13-42.802156.parquet - 分割:
latest - 路径:
**/details_harness|drop|3_2023-10-12T16-13-42.802156.parquet
- 分割:
-
harness_gsm8k_5
- 分割:
2023_10_12T16_13_42.802156 - 路径:
**/details_harness|gsm8k|5_2023-10-12T16-13-42.802156.parquet - 分割:
latest - 路径:
**/details_harness|gsm8k|5_2023-10-12T16-13-42.802156.parquet
- 分割:
-
harness_hellaswag_10
- 分割:
2023_07_18T16_42_29.519016 - 路径:
**/details_harness|hellaswag|10_2023-07-18T16:42:29.519016.parquet - 分割:
latest - 路径:
**/details_harness|hellaswag|10_2023-07-18T16:42:29.519016.parquet
- 分割:
-
harness_hendrycksTest_5
- 分割:
2023_07_18T16_42_29.519016 - 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-anatomy|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-astronomy|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-college_biology|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-college_chemistry|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-college_computer_science|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-college_mathematics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-college_medicine|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-college_physics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-computer_security|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-econometrics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-formal_logic|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-global_facts|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_biology|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_geography|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_physics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-human_aging|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-human_sexuality|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-international_law|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-jurisprudence|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-machine_learning|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-management|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-marketing|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-medical_genetics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-miscellaneous|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-moral_disputes|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-nutrition|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-philosophy|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-prehistory|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-professional_accounting|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-professional_law|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-professional_medicine|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-professional_psychology|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-public_relations|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-security_studies|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-sociology|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-us_foreign_policy|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-virology|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-world_religions|5_2023-07-18T16:42:29.519016.parquet
- 分割:
latest - 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-anatomy|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-astronomy|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-business_ethics|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-18T16:42:29.519016.parquet**/details_harness|hendrycksTest-college_biology|5_2023-07-18T16:42:29.519016.parquet- `**/details_harness|hendrycksTest-college_chemistry|5_202
- 分割:



