open-llm-leaderboard/details_Undi95__CodeEngine

该数据集是在Open LLM Leaderboard上对模型Undi95/CodeEngine进行评估时自动生成的。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，分割名称由运行的时间戳命名。train分割始终指向最新的结果。此外，名为results的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还提供了如何使用Python代码加载运行中的详细信息的示例，并包含了特定运行的最新结果。

This dataset was automatically generated during the evaluation of the model Undi95/CodeEngine on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to an evaluated task. The dataset is generated from two runs, where each run is represented as a specific split under each configuration, with the split name named after the run's timestamp. The train split always points to the most recent results. Additionally, the configuration named "results" stores aggregated results across all runs, which are utilized to calculate and display the aggregated metrics on the Open LLM Leaderboard. The README also provides examples of how to load detailed run information using Python code, and includes the latest results for specific runs.

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 Undi95/CodeEngine 进行评估运行期间自动创建的，用于 Open LLM Leaderboard。

数据集结构

• 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 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_Undi95__CodeEngine", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-25T07:16:21.496689 运行 的最新结果： python { "all": { "em": 0.31008808724832215, "em_stderr": 0.004736734191590966, "f1": 0.4059154781879224, "f1_stderr": 0.004594505528583743, "acc": 0.38050967793280527, "acc_stderr": 0.00780116508471732 }, "harness|drop|3": { "em": 0.31008808724832215, "em_stderr": 0.004736734191590966, "f1": 0.4059154781879224, "f1_stderr": 0.004594505528583743 }, "harness|gsm8k|5": { "acc": 0.015163002274450341, "acc_stderr": 0.0033660229497263702 }, "harness|winogrande|5": { "acc": 0.7458563535911602, "acc_stderr": 0.012236307219708269 } }

配置详情

以下是部分配置及其数据文件路径：

• config_name: harness_arc_challenge_25

  • split: 2023_09_12T11_51_31.235775
    • path: **/details_harness|arc:challenge|25_2023-09-12T11-51-31.235775.parquet
  • split: latest
    • path: **/details_harness|arc:challenge|25_2023-09-12T11-51-31.235775.parquet

• config_name: harness_drop_3

  • split: 2023_10_25T07_16_21.496689
    • path: **/details_harness|drop|3_2023-10-25T07-16-21.496689.parquet
  • split: latest
    • path: **/details_harness|drop|3_2023-10-25T07-16-21.496689.parquet

• config_name: harness_gsm8k_5

  • split: 2023_10_25T07_16_21.496689
    • path: **/details_harness|gsm8k|5_2023-10-25T07-16-21.496689.parquet
  • split: latest
    • path: **/details_harness|gsm8k|5_2023-10-25T07-16-21.496689.parquet

• config_name: harness_hellaswag_10

  • split: 2023_09_12T11_51_31.235775
    • path: **/details_harness|hellaswag|10_2023-09-12T11-51-31.235775.parquet
  • split: latest
    • path: **/details_harness|hellaswag|10_2023-09-12T11-51-31.235775.parquet

• config_name: harness_hendrycksTest_5

  • split: 2023_09_12T11_51_31.235775
    • path:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-12T11-51-31.235775.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-09-12T11-51-31.235775.parquet
      • ...
  • split: latest
    • path:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-12T11-51-31.235775.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-09-12T11-51-31.235775.parquet
      • ...