open-llm-leaderboard/details_grimjim__zephyr-wizard-kuno-royale-BF16-merge-7B

该数据集是在模型[grimjim/zephyr-wizard-kuno-royale-BF16-merge-7B](https://huggingface.co/grimjim/zephyr-wizard-kuno-royale-BF16-merge-7B)的评估运行期间自动创建的，用于在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行评估。数据集由63个配置组成，每个配置对应一个评估任务。数据集由2次运行生成，每次运行的结果作为特定配置中的一个分割，分割名称使用运行的时间戳。"train"分割始终指向最新结果。此外，"results"配置存储了所有运行的聚合结果，并用于计算和显示[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/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 grimjim/zephyr-wizard-kuno-royale-BF16-merge-7B

数据集描述

• dataset_summary: 该数据集是在评估模型grimjim/zephyr-wizard-kuno-royale-BF16-merge-7B在Open LLM Leaderboard上的运行过程中自动创建的。

数据集组成

• 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2次运行中创建，每次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命名。
• 特殊配置: “results”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。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_grimjim__zephyr-wizard-kuno-royale-BF16-merge-7B", "harness_winogrande_5", split="train")

最新结果

• 结果来源: 来自2024-04-23T10:25:17.043076的运行结果。
• 结果内容: 包含多个任务的评估结果，如准确率(acc)、标准误差(acc_stderr)等。

数据集配置详情

配置列表

1. harness_arc_challenge_25
2. harness_gsm8k_5
3. harness_hellaswag_10
4. harness_hendrycksTest_5

数据文件路径

• harness_arc_challenge_25:

  • split: 2024_04_19T21_59_19.193146
    • path: /details_harness|arc:challenge|25_2024-04-19T21-59-19.193146.parquet
  • split: 2024_04_23T10_25_17.043076
    • path: /details_harness|arc:challenge|25_2024-04-23T10-25-17.043076.parquet
  • split: latest
    • path: /details_harness|arc:challenge|25_2024-04-23T10-25-17.043076.parquet

• harness_gsm8k_5:

  • split: 2024_04_19T21_59_19.193146
    • path: /details_harness|gsm8k|5_2024-04-19T21-59-19.193146.parquet
  • split: 2024_04_23T10_25_17.043076
    • path: /details_harness|gsm8k|5_2024-04-23T10-25-17.043076.parquet
  • split: latest
    • path: /details_harness|gsm8k|5_2024-04-23T10-25-17.043076.parquet

• harness_hellaswag_10:

  • split: 2024_04_19T21_59_19.193146
    • path: /details_harness|hellaswag|10_2024-04-19T21-59-19.193146.parquet
  • split: 2024_04_23T10_25_17.043076
    • path: /details_harness|hellaswag|10_2024-04-23T10-25-17.043076.parquet
  • split: latest
    • path: /details_harness|hellaswag|10_2024-04-23T10-25-17.043076.parquet

• harness_hendrycksTest_5:

  • split: 2024_04_19T21_59_19.193146
    • path: /details_harness|hendrycksTest-abstract_algebra|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-anatomy|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-astronomy|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-business_ethics|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-clinical_knowledge|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-college_biology|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-college_chemistry|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-college_computer_science|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-college_mathematics|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-college_medicine|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-college_physics|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-computer_security|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-conceptual_physics|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-econometrics|5_2024-04-19T21-59-19.193146.parquet
    • path: /details_harness|hendrycksTest-electrical_engineering|5_2024-04-19T21-59-19.193146.parquet