open-llm-leaderboard/details_databricks__dolly-v2-3b

该数据集是在模型databricks/dolly-v2-3b在Open LLM Leaderboard上的评估运行期间自动创建的。它由64个配置组成，每个配置对应一个评估任务。数据集由3次运行生成，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。可以使用Python中的datasets库加载该数据集。

This dataset was automatically generated during the evaluation run of the model databricks/dolly-v2-3b on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to a single evaluation task. The dataset is produced from three runs, where each run is represented as a specific split under each configuration. The train split always references the most up-to-date results. An additional configuration titled `results` stores the aggregated results across all runs, which are utilized to compute and display the aggregate metrics on the Open LLM Leaderboard. This dataset can be loaded using the datasets library in Python.

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 databricks/dolly-v2-3b 进行评估运行期间自动创建的，用于 Open LLM Leaderboard 的评估。

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 3 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_databricks__dolly-v2-3b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-15T08:28:44.127308 运行的最新结果： python { "all": { "em": 0.001572986577181208, "em_stderr": 0.00040584511324177506, "f1": 0.05171245805369139, "f1_stderr": 0.0012518561178042446, "acc": 0.30246569325856754, "acc_stderr": 0.008311459829200955 }, "harness|drop|3": { "em": 0.001572986577181208, "em_stderr": 0.00040584511324177506, "f1": 0.05171245805369139, "f1_stderr": 0.0012518561178042446 }, "harness|gsm8k|5": { "acc": 0.01061410159211524, "acc_stderr": 0.002822713322387704 }, "harness|winogrande|5": { "acc": 0.5943172849250198, "acc_stderr": 0.013800206336014207 } }

配置详情

• harness_arc_challenge_25

  • 分割：2023_07_19T15_08_45.552470, latest
  • 路径：**/details_harness|arc:challenge|25_2023-07-19T15:08:45.552470.parquet

• harness_drop_3

  • 分割：2023_10_12T19_06_58.415048, 2023_10_15T08_28_44.127308, latest
  • 路径：**/details_harness|drop|3_2023-10-12T19-06-58.415048.parquet, **/details_harness|drop|3_2023-10-15T08-28-44.127308.parquet

• harness_gsm8k_5

  • 分割：2023_10_12T19_06_58.415048, 2023_10_15T08_28_44.127308, latest
  • 路径：**/details_harness|gsm8k|5_2023-10-12T19-06-58.415048.parquet, **/details_harness|gsm8k|5_2023-10-15T08-28-44.127308.parquet

• harness_hellaswag_10

  • 分割：2023_07_19T15_08_45.552470, latest
  • 路径：**/details_harness|hellaswag|10_2023-07-19T15:08:45.552470.parquet

• harness_hendrycksTest_5

  • 分割：2023_07_19T15_08_45.552470, latest
  • 路径：多个路径，例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:08:45.552470.parquet

• 其他配置

  • 包括 harness_hendrycksTest_abstract_algebra_5, harness_hendrycksTest_anatomy_5, harness_hendrycksTest_astronomy_5 等，每个配置都有相应的分割和路径。