open-llm-leaderboard/details_jondurbin__airoboros-l2-13b-gpt4-1.4.1
收藏数据集概述
数据集简介
该数据集是在对模型 jondurbin/airoboros-l2-13b-gpt4-1.4.1 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
数据集结构
- 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,每次运行作为一个特定的分割存储,分割名称使用运行的时间戳。
- "train" 分割始终指向最新的结果。
- 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 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_jondurbin__airoboros-l2-13b-gpt4-1.4.1", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-22T19:03:13.374959 运行 的最新结果: python { "all": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642965807, "f1": 0.07133494127516771, "f1_stderr": 0.0015039896976380969, "acc": 0.40148895416572666, "acc_stderr": 0.00972321783657909 }, "harness|drop|3": { "em": 0.0026216442953020135, "em_stderr": 0.0005236685642965807, "f1": 0.07133494127516771, "f1_stderr": 0.0015039896976380969 }, "harness|gsm8k|5": { "acc": 0.06974981046247157, "acc_stderr": 0.00701638957101385 }, "harness|winogrande|5": { "acc": 0.7332280978689818, "acc_stderr": 0.012430046102144331 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|arc:challenge|25_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-07-24T15:17:43.655120.parquet
-
harness_drop_3
- 分割:2023_10_22T19_03_13.374959
- 路径:
**/details_harness|drop|3_2023-10-22T19-03-13.374959.parquet - 分割:latest
- 路径:
**/details_harness|drop|3_2023-10-22T19-03-13.374959.parquet
-
harness_gsm8k_5
- 分割:2023_10_22T19_03_13.374959
- 路径:
**/details_harness|gsm8k|5_2023-10-22T19-03-13.374959.parquet - 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-22T19-03-13.374959.parquet
-
harness_hellaswag_10
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hellaswag|10_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T15:17:43.655120.parquet等 40 个文件 - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T15:17:43.655120.parquet等 40 个文件
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_07_24T15_17_43.655120
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-24T15:17:43.655120.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-24T15:17:43.655120.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_07_24T15_17_43.655120
- 路径:`**/details_harness|hendrycksTest-electrical_engineering|5_20



