open-llm-leaderboard-old/details_NekoPunchBBB__Llama-2-13b-hf_Open-Platypus
收藏数据集概述
数据集简介
该数据集是在评估模型 NekoPunchBBB/Llama-2-13b-hf_Open-Platypus 在 Open LLM Leaderboard 上的自动创建的。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
- 配置数量:64 个配置
- 运行次数:数据集从 2 次运行中创建。每个运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。
- 最新结果:"train" 分片总是指向最新的结果。
- 结果汇总:一个额外的配置 "results" 存储所有运行的汇总结果,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_NekoPunchBBB__Llama-2-13b-hf_Open-Platypus", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-28T23:55:29.524806 运行的最新结果: python { "all": { "em": 0.0017827181208053692, "em_stderr": 0.00043200973460388544, "f1": 0.05985213926174496, "f1_stderr": 0.0013641672120704657, "acc": 0.4325617395685546, "acc_stderr": 0.009923090021448928 }, "harness|drop|3": { "em": 0.0017827181208053692, "em_stderr": 0.00043200973460388544, "f1": 0.05985213926174496, "f1_stderr": 0.0013641672120704657 }, "harness|gsm8k|5": { "acc": 0.09401061410159212, "acc_stderr": 0.00803881981887246 }, "harness|winogrande|5": { "acc": 0.771112865035517, "acc_stderr": 0.011807360224025398 } }
配置详情
以下是部分配置的详细信息:
-
harness_arc_challenge_25
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|arc:challenge|25_2023-09-11T16-11-41.270351.parquet
-
harness_drop_3
- 分片:2023_10_28T23_55_29.524806, latest
- 路径:
**/details_harness|drop|3_2023-10-28T23-55-29.524806.parquet
-
harness_gsm8k_5
- 分片:2023_10_28T23_55_29.524806, latest
- 路径:
**/details_harness|gsm8k|5_2023-10-28T23-55-29.524806.parquet
-
harness_hellaswag_10
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hellaswag|10_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:多个路径,例如
**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_abstract_algebra_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_anatomy_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_astronomy_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_business_ethics_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_college_biology_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_college_medicine_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_college_physics_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_computer_security_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_econometrics_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-11T16-11-41.270351.parquet
-
harness_hendrycksTest_elementary_mathematics_5
- 分片:2023_09_11T16_11_41.270351, latest
- 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-11T16-11-41.270351.parquet
以上是数据集的部分配置详情,完整配置信息请参考原始文档。



