open-llm-leaderboard-old/details_nbeerbower__flammen

该数据集是在Open LLM Leaderboard上对模型nbeerbower/flammen进行评估时自动创建的。数据集由63个配置组成，每个配置对应一个评估任务。数据集是从1次运行中创建的，每次运行作为每个配置中的一个特定分割，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。文件还提供了如何使用Python代码加载运行细节的示例，并包含了特定运行的最新结果。

原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 nbeerbower/flammen 进行评估运行期间自动创建的，用于 Open LLM Leaderboard。

数据集结构

• 配置数量：63个配置，每个配置对应一个评估任务。
• 数据来源：数据集从1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" 分割总是指向最新的结果。
• 汇总结果：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_nbeerbower__flammen", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果 的摘要：

python { "all": { "acc": 0.6507639681694609, "acc_stderr": 0.03207791775155649, "acc_norm": 0.6512277255332111, "acc_norm_stderr": 0.03273350205907639, "mc1": 0.45777233782129745, "mc1_stderr": 0.01744096571248212, "mc2": 0.6301596481410238, "mc2_stderr": 0.015329932733848023 }, "harness|arc:challenge|25": { "acc": 0.6569965870307167, "acc_stderr": 0.013872423223718166, "acc_norm": 0.681740614334471, "acc_norm_stderr": 0.013611993916971453 }, "harness|hellaswag|10": { "acc": 0.6946823341963753, "acc_stderr": 0.004596006250433552, "acc_norm": 0.8706432981477793, "acc_norm_stderr": 0.0033490845685472596 }, # 更多任务结果... }

配置详情

• config_name: harness_arc_challenge_25

  • split: 2024_02_29T18_59_23.861321
    • path: **/details_harness|arc:challenge|25_2024-02-29T18-59-23.861321.parquet
  • split: latest
    • path: **/details_harness|arc:challenge|25_2024-02-29T18-59-23.861321.parquet

• config_name: harness_gsm8k_5

  • split: 2024_02_29T18_59_23.861321
    • path: **/details_harness|gsm8k|5_2024-02-29T18-59-23.861321.parquet
  • split: latest
    • path: **/details_harness|gsm8k|5_2024-02-29T18-59-23.861321.parquet

• config_name: harness_hellaswag_10

  • split: 2024_02_29T18_59_23.861321
    • path: **/details_harness|hellaswag|10_2024-02-29T18-59-23.861321.parquet
  • split: latest
    • path: **/details_harness|hellaswag|10_2024-02-29T18-59-23.861321.parquet

• config_name: harness_hendrycksTest_5

  • split: 2024_02_29T18_59_23.861321
    • path:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2024-02-29T18-59-23.861321.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-29T18-59-23.861321.parquet

      更多路径...

}