open-llm-leaderboard-old/details_shibing624__chinese-alpaca-plus-13b-hf

该数据集是在评估模型shibing624/chinese-alpaca-plus-13b-hf运行期间自动创建的，包含64个配置，每个配置对应一个评估任务。数据集包含来自2次运行的数据，每次运行作为一个特定分割，以运行的时间戳命名。train分割始终指向最新结果。额外的results配置存储了运行的所有聚合结果，用于在Open LLM Leaderboard上计算和显示聚合指标。数据集结构化，具有针对不同任务的各种配置，并且每个配置都有与不同分割相关的特定数据文件。

原始信息汇总

数据集概述

该数据集是在评估模型 shibing624/chinese-alpaca-plus-13b-hf 在 Open LLM Leaderboard 上的运行过程中自动创建的。

数据集组成

• 数据集包含 64 个配置，每个配置对应一个评估任务。
• 数据集由 2 次运行创建，每次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。
• "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_shibing624__chinese-alpaca-plus-13b-hf", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果来自 run 2023-09-22T22:17:42.526571 的摘要： python { "all": { "em": 0.21413590604026847, "em_stderr": 0.004201052628405755, "f1": 0.28453125000000146, "f1_stderr": 0.004252615634369811, "acc": 0.3859100763356038, "acc_stderr": 0.00806531916338966 }, "harness|drop|3": { "em": 0.21413590604026847, "em_stderr": 0.004201052628405755, "f1": 0.28453125000000146, "f1_stderr": 0.004252615634369811 }, "harness|gsm8k|5": { "acc": 0.02122820318423048, "acc_stderr": 0.003970449129848635 }, "harness|winogrande|5": { "acc": 0.7505919494869772, "acc_stderr": 0.012160189196930687 } }

配置详情

以下是数据集的配置详情：

• harness_arc_challenge_25

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|arc:challenge|25_2023-07-19T19:20:40.370845.parquet

• harness_drop_3

  • 分割: 2023_09_22T22_17_42.526571, latest
  • 路径: **/details_harness|drop|3_2023-09-22T22-17-42.526571.parquet

• harness_gsm8k_5

  • 分割: 2023_09_22T22_17_42.526571, latest
  • 路径: **/details_harness|gsm8k|5_2023-09-22T22-17-42.526571.parquet

• harness_hellaswag_10

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hellaswag|10_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: 多个路径，例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_abstract_algebra_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_anatomy_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_astronomy_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_business_ethics_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_clinical_knowledge_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_college_biology_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_college_chemistry_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_college_computer_science_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_college_mathematics_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_college_medicine_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_college_physics_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_computer_security_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_conceptual_physics_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_econometrics_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_electrical_engineering_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T19:20:40.370845.parquet

• harness_hendrycksTest_elementary_mathematics_5

  • 分割: 2023_07_19T19_20_40.370845, latest
  • 路径: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T19:20:40.370845.parquet

以上是数据集的详细概述，包括数据集的组成、加载示例、最新结果和配置详情。