open-llm-leaderboard-old/details_golaxy__gogpt-7b

该数据集是在模型 golaxy/gogpt-7b 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成，每个配置对应一个被评估的任务。数据集是从 2 次运行中生成的，每次运行在每个配置中表示为特定的分割，分割名称由运行的时间戳命名。train 分割始终指向最新的结果。一个名为 results 的额外配置存储了所有运行的聚合结果，这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个示例，展示了如何使用 `datasets` 库中的 `load_dataset` 函数加载运行中的详细信息。

原始信息汇总

数据集概述

数据集来源

该数据集是在对模型 golaxy/gogpt-7b 进行评估运行期间自动创建的，评估结果展示在 Open LLM Leaderboard 上。

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。
• 运行次数：数据集从2次运行中创建。每个运行在每个配置中作为一个特定的分割存在，分割名称使用运行的时间戳。
• 训练分割："train" 分割始终指向最新的结果。
• 结果配置：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_golaxy__gogpt-7b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-16T19:51:01.588923 运行 的最新结果： python { "all": { "em": 0.27936241610738255, "em_stderr": 0.0045949638030960225, "f1": 0.3275786493288599, "f1_stderr": 0.004570156166276547, "acc": 0.358332440746709, "acc_stderr": 0.008331639351329504 }, "harness|drop|3": { "em": 0.27936241610738255, "em_stderr": 0.0045949638030960225, "f1": 0.3275786493288599, "f1_stderr": 0.004570156166276547 }, "harness|gsm8k|5": { "acc": 0.018953752843062926, "acc_stderr": 0.0037560783410314704 }, "harness|winogrande|5": { "acc": 0.6977111286503551, "acc_stderr": 0.012907200361627538 } }

配置详情

• harness_arc_challenge_25

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|arc:challenge|25_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|arc:challenge|25_2023-07-24T11:32:55.056664.parquet

• harness_drop_3

  • 分割：2023_10_16T19_51_01.588923
  • 路径：**/details_harness|drop|3_2023-10-16T19-51-01.588923.parquet
  • 分割：latest
  • 路径：**/details_harness|drop|3_2023-10-16T19-51-01.588923.parquet

• harness_gsm8k_5

  • 分割：2023_10_16T19_51_01.588923
  • 路径：**/details_harness|gsm8k|5_2023-10-16T19-51-01.588923.parquet
  • 分割：latest
  • 路径：**/details_harness|gsm8k|5_2023-10-16T19-51-01.588923.parquet

• harness_hellaswag_10

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hellaswag|10_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hellaswag|10_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T11:32:55.056664.parquet 等
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T11:32:55.056664.parquet 等

• harness_hendrycksTest_abstract_algebra_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_anatomy_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-anatomy|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-anatomy|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_astronomy_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-astronomy|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-astronomy|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_business_ethics_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-business_ethics|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-business_ethics|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_clinical_knowledge_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_college_biology_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-college_biology|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-college_biology|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_college_chemistry_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_college_computer_science_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_college_mathematics_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_college_medicine_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-college_medicine|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-college_medicine|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_college_physics_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-college_physics|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-college_physics|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_computer_security_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-computer_security|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-computer_security|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_conceptual_physics_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_econometrics_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-econometrics|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-econometrics|5_2023-07-24T11:32:55.056664.parquet

• harness_hendrycksTest_electrical_engineering_5

  • 分割：2023_07_24T11_32_55.056664
  • 路径：**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-24T11:32:55.056664.parquet
  • 分割：latest
  • 路径：`**/details_harness|hendrycksTest-electrical_engineering|5