five

open-llm-leaderboard/details_togethercomputer__RedPajama-INCITE-Base-7B-v0.1

收藏
Hugging Face2023-10-23 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_togethercomputer__RedPajama-INCITE-Base-7B-v0.1
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-Base-7B-v0.1在Open LLM Leaderboard上进行评估时自动创建的。数据集由64个配置组成,每个配置对应一个被评估的任务。它包含2次运行的数据,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还提供了如何使用Hugging Face datasets库加载数据集的示例。
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-Base-7B-v0.1 的过程中自动创建的,用于 Open LLM Leaderboard 上的评估。

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在每个配置中找到,以运行的时间戳命名的特定分割形式存储。
  • 每个配置中的 "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,用于计算并在 Open LLM Leaderboard 上显示聚合指标。

数据加载示例

以下是一个加载特定运行结果的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_togethercomputer__RedPajama-INCITE-Base-7B-v0.1", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-23T19:08:48.233759 运行 的最新结果: python { "all": { "em": 0.0010486577181208054, "em_stderr": 0.00033145814652192694, "f1": 0.05110738255033561, "f1_stderr": 0.0012343063700893503, "acc": 0.3445825177884037, "acc_stderr": 0.008314908287260184 }, "harness|drop|3": { "em": 0.0010486577181208054, "em_stderr": 0.00033145814652192694, "f1": 0.05110738255033561, "f1_stderr": 0.0012343063700893503 }, "harness|gsm8k|5": { "acc": 0.01592115238817286, "acc_stderr": 0.0034478192723889985 }, "harness|winogrande|5": { "acc": 0.6732438831886346, "acc_stderr": 0.013181997302131368 } }

配置详情

以下是部分配置及其数据文件路径:

  • config_name: harness_arc_challenge_25

    • split: 2023_07_19T16_33_56.917496
      • path: **/details_harness|arc:challenge|25_2023-07-19T16:33:56.917496.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-07-19T16:33:56.917496.parquet
  • config_name: harness_drop_3

    • split: 2023_10_23T19_08_48.233759
      • path: **/details_harness|drop|3_2023-10-23T19-08-48.233759.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-10-23T19-08-48.233759.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_10_23T19_08_48.233759
      • path: **/details_harness|gsm8k|5_2023-10-23T19-08-48.233759.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-10-23T19-08-48.233759.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_07_19T16_33_56.917496
      • path: **/details_harness|hellaswag|10_2023-07-19T16:33:56.917496.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-07-19T16:33:56.917496.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_07_19T16_33_56.917496
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-management|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T16:33:56.917496.parquet
        • **/details_harness|hendrycks
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