open-llm-leaderboard/details_Doctor-Shotgun__CalliopeDS-L2-13B
收藏数据集概述
数据集简介
该数据集是在评估模型 Doctor-Shotgun/CalliopeDS-L2-13B 在 Open LLM Leaderboard 上的自动创建的。
数据集结构
- 配置数量:64个配置,每个配置对应一个评估任务。
- 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。
- 训练分片:"train" 分片始终指向最新的结果。
- 结果配置: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_Doctor-Shotgun__CalliopeDS-L2-13B", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-26T04:36:21.549191 运行的最新结果: python { "all": { "em": 0.02307046979865772, "em_stderr": 0.0015374446489046481, "f1": 0.08979446308724821, "f1_stderr": 0.0020360011017500185, "acc": 0.4351997070321265, "acc_stderr": 0.010043960065261932 }, "harness|drop|3": { "em": 0.02307046979865772, "em_stderr": 0.0015374446489046481, "f1": 0.08979446308724821, "f1_stderr": 0.0020360011017500185 }, "harness|gsm8k|5": { "acc": 0.10007581501137225, "acc_stderr": 0.008266274528685632 }, "harness|winogrande|5": { "acc": 0.7703235990528808, "acc_stderr": 0.011821645601838232 } }
配置详情
以下是数据集的部分配置详情:
-
harness_arc_challenge_25
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|arc:challenge|25_2023-09-18T14-00-51.912601.parquet
-
harness_drop_3
- 分片:2023_10_26T04_36_21.549191, latest
- 路径:**/details_harness|drop|3_2023-10-26T04-36-21.549191.parquet
-
harness_gsm8k_5
- 分片:2023_10_26T04_36_21.549191, latest
- 路径:**/details_harness|gsm8k|5_2023-10-26T04-36-21.549191.parquet
-
harness_hellaswag_10
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hellaswag|10_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:多个路径,包括但不限于 **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_abstract_algebra_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_anatomy_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_astronomy_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_business_ethics_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_college_biology_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_college_medicine_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_college_physics_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_computer_security_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_econometrics_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-18T14-00-51.912601.parquet
-
harness_hendrycksTest_elementary_mathematics_5
- 分片:2023_09_18T14_00_51.912601, latest
- 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-18T14-00-51.912601.parquet
以上是数据集的部分配置详情,完整配置详情请参考原始文档。



