open-llm-leaderboard-old/details_CobraMamba__mamba-gpt-3b

该数据集是在Open LLM Leaderboard上对模型CobraMamba/mamba-gpt-3b进行评估时自动创建的。数据集由64个配置组成，每个配置对应一个特定的评估任务。数据集由2次运行生成，每次运行在每个配置中表示为特定的分割，分割名称使用运行的时间戳。train分割始终指向最新的结果。此外，还有一个名为results的配置，存储了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。README还提供了如何使用`datasets`库中的`load_dataset`函数加载数据集的示例。

原始信息汇总

数据集概述

该数据集是在对模型 CobraMamba/mamba-gpt-3b 进行评估运行期间自动创建的，用于 Open LLM Leaderboard。

数据集组成

• 数据集由 64 个配置组成，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 Open LLM Leaderboard 上的聚合指标。

最新结果

• 最新结果来自 2023-10-14T18:11:22.900658 运行，包含多个任务的结果。

配置详情

• harness_arc_challenge_25

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|arc:challenge|25_2023-07-19T15:20:46.724343.parquet

• harness_drop_3

  • 分割：2023_10_14T18_11_22.900658, latest
  • 路径：**/details_harness|drop|3_2023-10-14T18-11-22.900658.parquet

• harness_gsm8k_5

  • 分割：2023_10_14T18_11_22.900658, latest
  • 路径：**/details_harness|gsm8k|5_2023-10-14T18-11-22.900658.parquet

• harness_hellaswag_10

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hellaswag|10_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：多个路径，例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_abstract_algebra_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_anatomy_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-anatomy|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_astronomy_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-astronomy|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_business_ethics_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_clinical_knowledge_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_college_biology_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-college_biology|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_college_chemistry_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_college_computer_science_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_college_mathematics_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_college_medicine_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_college_physics_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-college_physics|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_computer_security_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-computer_security|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_conceptual_physics_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_econometrics_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-econometrics|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_electrical_engineering_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T15:20:46.724343.parquet

• harness_hendrycksTest_elementary_mathematics_5

  • 分割：2023_07_19T15_20_46.724343, latest
  • 路径：**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T15:20:46.724343.parquet