open-llm-leaderboard/details_yeontaek__Platypus2xOpenOrca-13B-LoRa-v2
收藏数据集概述
数据集简介
该数据集是在评估模型yeontaek/Platypus2xOpenOrca-13B-LoRa-v2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
- 数据来源:从2次运行中创建,每个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。
- 最新结果:"train"分割始终指向最新结果。
- 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_yeontaek__Platypus2xOpenOrca-13B-LoRa-v2", "harness_winogrande_5", split="train")
最新结果
以下是最新结果来自2023-10-22T18:31:59.228991: python { "all": { "em": 0.012688758389261746, "em_stderr": 0.0011462418380586469, "f1": 0.09433724832214756, "f1_stderr": 0.002007935719484454, "acc": 0.38119991167880285, "acc_stderr": 0.0063665493907124155 }, "harness|drop|3": { "em": 0.012688758389261746, "em_stderr": 0.0011462418380586469, "f1": 0.09433724832214756, "f1_stderr": 0.002007935719484454 }, "harness|gsm8k|5": { "acc": 0.000758150113722517, "acc_stderr": 0.0007581501137225193 }, "harness|winogrande|5": { "acc": 0.7616416732438832, "acc_stderr": 0.011974948667702311 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|arc:challenge|25_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-23T06:03:44.232629.parquet
-
harness_drop_3
- 分割:2023_10_22T18_31_59.228991
- 路径:
**/details_harness|drop|3_2023-10-22T18-31-59.228991.parquet - 分割:latest
- 路径:
**/details_harness|drop|3_2023-10-22T18-31-59.228991.parquet
-
harness_gsm8k_5
- 分割:2023_10_22T18_31_59.228991
- 路径:
**/details_harness|gsm8k|5_2023-10-22T18-31-59.228991.parquet - 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-22T18-31-59.228991.parquet
-
harness_hellaswag_10
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hellaswag|10_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_5
- 分割:2023_08_23T06_03_44.232629
- 路径:多个路径,例如
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-23T06:03:44.232629.parquet等 - 分割:latest
- 路径:多个路径,例如
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-23T06:03:44.232629.parquet等
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_08_23T06_03_44.232629
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-23T06:03:44.232629.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-23T06:03:44.232629.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_08_23T06_03_44.232629
- 路径:`**/details_harness|hendrycksTest-electrical_



