open-llm-leaderboard/details_HyperbeeAI__Tulpar-7b-v0

该数据集是在Open LLM Leaderboard上对模型HyperbeeAI/Tulpar-7b-v0进行评估时自动创建的。数据集由64个配置组成，每个配置对应一个被评估的任务。数据集由2次运行创建，每次运行在每个配置中表示为特定的分割。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，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`load_dataset`函数加载运行中的详细信息的示例。

This dataset was automatically created when evaluating the model HyperbeeAI/Tulpar-7b-v0 on the Open LLM Leaderboard. It consists of 64 configurations, each corresponding to one evaluated task. The dataset is generated from two runs, where each run is represented as a specific split within each configuration. The `train` split always points to the most recent results. Additionally, a configuration named `results` stores the aggregated results of all runs, which are used to calculate and display the aggregated metrics on the Open LLM Leaderboard. The README also provides examples of how to use the `load_dataset` function to load detailed information from the runs.

原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 HyperbeeAI/Tulpar-7b-v0 在 Open LLM Leaderboard 上的运行过程中自动创建的。

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_HyperbeeAI__Tulpar-7b-v0", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-16T17:05:33.641696 运行的最新结果： python { "all": { "em": 0.3200503355704698, "em_stderr": 0.004777351284269766, "f1": 0.39745910234899495, "f1_stderr": 0.004660867839676267, "acc": 0.38302318192072277, "acc_stderr": 0.00841750512181253 }, "harness|drop|3": { "em": 0.3200503355704698, "em_stderr": 0.004777351284269766, "f1": 0.39745910234899495, "f1_stderr": 0.004660867839676267 }, "harness|gsm8k|5": { "acc": 0.027293404094010616, "acc_stderr": 0.004488095380209751 }, "harness|winogrande|5": { "acc": 0.7387529597474349, "acc_stderr": 0.012346914863415308 } }

配置详情

以下是部分配置的详细信息：

• harness_arc_challenge_25

  • 分割：2023_08_26T12_16_04.808575
  • 路径：**/details_harness|arc:challenge|25_2023-08-26T12:16:04.808575.parquet
  • 分割：latest
  • 路径：**/details_harness|arc:challenge|25_2023-08-26T12:16:04.808575.parquet

• harness_drop_3

  • 分割：2023_09_16T17_05_33.641696
  • 路径：**/details_harness|drop|3_2023-09-16T17-05-33.641696.parquet
  • 分割：latest
  • 路径：**/details_harness|drop|3_2023-09-16T17-05-33.641696.parquet

• harness_gsm8k_5

  • 分割：2023_09_16T17_05_33.641696
  • 路径：**/details_harness|gsm8k|5_2023-09-16T17-05-33.641696.parquet
  • 分割：latest
  • 路径：**/details_harness|gsm8k|5_2023-09-16T17-05-33.641696.parquet

• harness_hellaswag_10

  • 分割：2023_08_26T12_16_04.808575
  • 路径：**/details_harness|hellaswag|10_2023-08-26T12:16:04.808575.parquet
  • 分割：latest
  • 路径：**/details_harness|hellaswag|10_2023-08-26T12:16:04.808575.parquet

• harness_hendrycksTest_5

  • 分割：2023_08_26T12_16_04.808575
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-26T12:16:04.808575.parquet 等
  • 分割：latest
  • 路径：**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-26T12:16:04.808575.parquet 等