open-llm-leaderboard-old/details_llama-anon__instruct-13b
收藏数据集概述
数据集名称
- Evaluation run of llama-anon/instruct-13b
数据集组成
- 该数据集包含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。
数据集结构
- 数据集包含多个配置,每个配置有不同的数据文件和分割。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_llama-anon__instruct-13b", "harness_winogrande_5", split="train")
最新结果
- 最新结果来自2023-09-17T02:24:06.962063运行,包含多个任务的评估指标。
配置详情
-
harness_arc_challenge_25
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|arc:challenge|25_2023-07-19T18:48:36.816075.parquet
-
harness_drop_3
- 分割:2023_09_17T02_24_06.962063, latest
- 路径:
**/details_harness|drop|3_2023-09-17T02-24-06.962063.parquet
-
harness_gsm8k_5
- 分割:2023_09_17T02_24_06.962063, latest
- 路径:
**/details_harness|gsm8k|5_2023-09-17T02-24-06.962063.parquet
-
harness_hellaswag_10
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hellaswag|10_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:多个文件路径,例如
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T18:48:36.816075.parquet
-
harness_hendrycksTest_elementary_mathematics_5
- 分割:2023_07_19T18_48_36.816075, latest
- 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T18:48:36.816075.parquet



