five

open-llm-leaderboard/details_togethercomputer__RedPajama-INCITE-Instruct-3B-v1

收藏
Hugging Face2023-10-18 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_togethercomputer__RedPajama-INCITE-Instruct-3B-v1
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型 togethercomputer/RedPajama-INCITE-Instruct-3B-v1 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。它由 2 次运行创建,每次运行在每个配置中表示为特定的分割,使用运行的时间戳命名。train 分割始终指向最新的结果。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。README 还提供了一个示例,展示了如何使用 Hugging Face 的 datasets 库加载运行中的详细信息。

This dataset was automatically generated during the evaluation run of the model togethercomputer/RedPajama-INCITE-Instruct-3B-v1 on the Open LLM Leaderboard. The dataset comprises 64 configurations, each corresponding to one evaluated task. It was created through two runs, with each run represented as a specific split within each configuration and named using the run's timestamp. The 'train' split always points to the most recent results. An additional configuration named 'results' stores the aggregated results across all runs, which are utilized to compute and display the aggregate metrics on the Open LLM Leaderboard. The README also includes an example showing how to use the Hugging Face datasets library to load the detailed information from the runs.
提供机构:
open-llm-leaderboard
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 togethercomputer/RedPajama-INCITE-Instruct-3B-v1 的过程中自动创建的,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 2 次运行结果组成,每次运行的结果可以通过特定的分割(split)来访问,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_togethercomputer__RedPajama-INCITE-Instruct-3B-v1", "harness_winogrande_5", split="train")

最新结果

以下是来自 2023-10-18T03:35:35.026041 运行的最新结果: python { "all": { "em": 0.001363255033557047, "em_stderr": 0.0003778609196460735, "f1": 0.04506711409395986, "f1_stderr": 0.0011775010391852176, "acc": 0.32923850493036927, "acc_stderr": 0.00832289727752502 }, "harness|drop|3": { "em": 0.001363255033557047, "em_stderr": 0.0003778609196460735, "f1": 0.04506711409395986, "f1_stderr": 0.0011775010391852176 }, "harness|gsm8k|5": { "acc": 0.013646702047005308, "acc_stderr": 0.003195747075480785 }, "harness|winogrande|5": { "acc": 0.6448303078137332, "acc_stderr": 0.013450047479569256 } }

配置详情

以下是部分配置及其数据文件路径:

  • harness_arc_challenge_25

    • 分割: 2023_07_19T14_55_52.470090
      • 路径: **/details_harness|arc:challenge|25_2023-07-19T14:55:52.470090.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2023-07-19T14:55:52.470090.parquet
  • harness_drop_3

    • 分割: 2023_10_18T03_35_35.026041
      • 路径: **/details_harness|drop|3_2023-10-18T03-35-35.026041.parquet
    • 分割: latest
      • 路径: **/details_harness|drop|3_2023-10-18T03-35-35.026041.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_18T03_35_35.026041
      • 路径: **/details_harness|gsm8k|5_2023-10-18T03-35-35.026041.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2023-10-18T03-35-35.026041.parquet
  • harness_hellaswag_10

    • 分割: 2023_07_19T14_55_52.470090
      • 路径: **/details_harness|hellaswag|10_2023-07-19T14:55:52.470090.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2023-07-19T14:55:52.470090.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_07_19T14_55_52.470090
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-management|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-professional_medicine|5_2023-07-19T14:55:52.470090.parquet
        • **/details_harness|hendrycksTest-professional_psychology|5_2023-07-19T14:55:52.470090.parquet
        • `**/details_harness|hendrycksTest-public_relations|5_20
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