five

open-llm-leaderboard/details_oh-yeontaek__llama-2-70B-LoRA-assemble

收藏
Hugging Face2023-09-14 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_oh-yeontaek__llama-2-70B-LoRA-assemble
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型oh-yeontaek/llama-2-70B-LoRA-assemble时自动创建的,评估过程在Open LLM Leaderboard上进行。数据集包含61个配置,每个配置对应一个评估任务。数据集由1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,用于计算和显示Open LLM Leaderboard上的聚合指标。

This dataset was automatically created during the evaluation of the model oh-yeontaek/llama-2-70B-LoRA-assemble, which was conducted on the Open LLM Leaderboard. The dataset contains 61 configurations, each corresponding to one evaluation task. It is generated from a single run, where the results of each run act as a split under its specific configuration, and the split names are taken from the timestamp of the corresponding run. The train split always points to the latest evaluation result. Additionally, the "results" configuration stores the aggregated results of all runs, which are used to calculate and display the aggregated metrics on the Open LLM Leaderboard.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

该数据集是在评估模型oh-yeontaek/llama-2-70B-LoRA-assembleOpen LLM Leaderboard上的运行过程中自动创建的。

数据集组成

  • 数据集包含61个配置,每个配置对应一个评估任务。
  • 数据集由1次运行创建,每个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。
  • 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。

最新结果

  • 最新结果来自2023-09-14T11:41:03.022396的运行。
  • 结果包含多个任务的准确率(acc)、标准误差(acc_stderr)、归一化准确率(acc_norm)和归一化标准误差(acc_norm_stderr)等指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_oh-yeontaek__llama-2-70B-LoRA-assemble", "harness_truthfulqa_mc_0", split="train")

配置详情

  • config_name: harness_arc_challenge_25

    • 数据文件路径:
      • 分割: 2023_09_14T11_41_03.022396
        • **/details_harness|arc:challenge|25_2023-09-14T11-41-03.022396.parquet
      • 分割: latest
        • **/details_harness|arc:challenge|25_2023-09-14T11-41-03.022396.parquet
  • config_name: harness_hellaswag_10

    • 数据文件路径:
      • 分割: 2023_09_14T11_41_03.022396
        • **/details_harness|hellaswag|10_2023-09-14T11-41-03.022396.parquet
      • 分割: latest
        • **/details_harness|hellaswag|10_2023-09-14T11-41-03.022396.parquet
  • config_name: harness_hendrycksTest_5

    • 数据文件路径:
      • 分割: 2023_09_14T11_41_03.022396
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-14T11-41-03.022396.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-14T11-41-03.022396.parquet
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作