open-llm-leaderboard/details_Lajonbot__vicuna-7b-v1.5-PL-lora_unload
收藏数据集概述
数据集简介
该数据集是在评估模型 Lajonbot/vicuna-7b-v1.5-PL-lora_unload 在 Open LLM Leaderboard 上的运行过程中自动创建的。
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。
- 创建来源:数据集由2次运行创建。每次运行在每个配置中都有一个特定的分割,分割名称使用运行的时间戳。
- 最新结果:"train" 分割始终指向最新结果。
- 聚合结果: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_Lajonbot__vicuna-7b-v1.5-PL-lora_unload", "harness_winogrande_5", split="train")
最新结果
以下是 2023-09-23T03:39:03.666834 运行 的最新结果: python { "all": { "em": 0.0065016778523489934, "em_stderr": 0.0008230684297223919, "f1": 0.06541946308724841, "f1_stderr": 0.0015883719778429714, "acc": 0.3959174184839032, "acc_stderr": 0.009871427981667812 }, "harness|drop|3": { "em": 0.0065016778523489934, "em_stderr": 0.0008230684297223919, "f1": 0.06541946308724841, "f1_stderr": 0.0015883719778429714 }, "harness|gsm8k|5": { "acc": 0.07202426080363912, "acc_stderr": 0.007121147983537124 }, "harness|winogrande|5": { "acc": 0.7198105761641673, "acc_stderr": 0.012621707979798499 } }
配置详情
以下是数据集的配置详情:
-
harness_arc_challenge_25
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|arc:challenge|25_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-02T16:36:13.785976.parquet
-
harness_drop_3
- 分割:2023_09_23T03_39_03.666834
- 路径:
**/details_harness|drop|3_2023-09-23T03-39-03.666834.parquet - 分割:latest
- 路径:
**/details_harness|drop|3_2023-09-23T03-39-03.666834.parquet
-
harness_gsm8k_5
- 分割:2023_09_23T03_39_03.666834
- 路径:
**/details_harness|gsm8k|5_2023-09-23T03-39-03.666834.parquet - 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-09-23T03-39-03.666834.parquet
-
harness_hellaswag_10
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hellaswag|10_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-02T16:36:13.785976.parquet等 - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-02T16:36:13.785976.parquet等
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_08_02T16_36_13.785976
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-02T16:36:13.785976.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-02T16:36:13.785976.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_08_02T16_36_13.785976
- 路径:`**/details_harness|hendrycksTest-electrical_engineering|5



