open-llm-leaderboard/details_Josephgflowers__Tinyllama-1.5B-Cinder-Test-6

该数据集是在评估模型Josephgflowers/Tinyllama-1.5B-Cinder-Test-6时自动生成的，主要用于Open LLM Leaderboard的评估任务。数据集包含63个配置，每个配置对应一个评估任务。数据集由2次运行生成，每次运行的结果存储为特定的split，split的名称使用运行的时间戳。train split始终指向最新的结果。此外，数据集还包含一个名为results的配置，用于存储所有运行的聚合结果，并用于计算和显示Open LLM Leaderboard上的聚合指标。

原始信息汇总

数据集概述

数据集名称

• Pretty Name: Evaluation run of Josephgflowers/Tinyllama-1.5B-Cinder-Test-6

数据集创建

• Creation: Dataset automatically created during the evaluation run of the model Josephgflowers/Tinyllama-1.5B-Cinder-Test-6 on the Open LLM Leaderboard.

数据集组成

• Composition: Composed of 63 configurations, each corresponding to one of the evaluated tasks.
• Runs: Created from 2 runs. Each run can be found as a specific split in each configuration, named using the timestamp of the run.
• Splits: The "train" split always points to the latest results.

额外配置

• Configuration: "results" stores all the aggregated results of the run, used to compute and display the aggregated metrics on the Open LLM Leaderboard.

数据集加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Josephgflowers__Tinyllama-1.5B-Cinder-Test-6", "harness_winogrande_5", split="train")

最新结果

• Latest Results: From run 2024-04-16T20:50:26.222574, detailed metrics for various tasks are provided.

数据集详细配置

配置列表

1. harness_arc_challenge_25
   • Data Files: Includes splits for two timestamps and the latest.
2. harness_gsm8k_5
   • Data Files: Includes splits for two timestamps and the latest.
3. harness_hellaswag_10
   • Data Files: Includes splits for two timestamps and the latest.
4. harness_hendrycksTest_5
   • Data Files: Includes multiple splits across various tasks for two timestamps and the latest.