five

open-llm-leaderboard/details_TigerResearch__tigerbot-70b-base

收藏
Hugging Face2023-10-24 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_TigerResearch__tigerbot-70b-base
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对TigerResearch/tigerbot-70b-base模型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分割始终指向最新的结果。此外,还有一个名为results的配置,存储了所有运行的聚合结果,用于在Open LLM Leaderboard上计算和显示聚合指标。

This dataset was automatically created during the evaluation of the TigerResearch/tigerbot-70b-base model on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluation task. It is generated from 2 runs, with the results of each run stored as a dataset split under the corresponding configuration, where the split name uses the timestamp of the run. The 'train' split always points to the most recent results. In addition, there is a configuration named 'results' that stores the aggregated results of all runs, which is used to calculate and display aggregate metrics on the Open LLM Leaderboard.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 TigerResearch/tigerbot-70b-base 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

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。
  • 数据集从 2 次运行中创建,每次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。
  • 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_TigerResearch__tigerbot-70b-base", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-24T09:25:20.725516 运行的最新结果: python { "all": { "em": 0.4872063758389262, "em_stderr": 0.005118791512925044, "f1": 0.5244914010067125, "f1_stderr": 0.004935563924712029, "acc": 0.5897264974960701, "acc_stderr": 0.012277506705422794 }, "harness|drop|3": { "em": 0.4872063758389262, "em_stderr": 0.005118791512925044, "f1": 0.5244914010067125, "f1_stderr": 0.004935563924712029 }, "harness|gsm8k|5": { "acc": 0.3775587566338135, "acc_stderr": 0.013353150666358539 }, "harness|winogrande|5": { "acc": 0.8018942383583267, "acc_stderr": 0.011201862744487047 } }

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_09_13T01_25_14.196261
      • path: **/details_harness|arc:challenge|25_2023-09-13T01-25-14.196261.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-09-13T01-25-14.196261.parquet
  • config_name: harness_drop_3

    • split: 2023_10_24T09_25_20.725516
      • path: **/details_harness|drop|3_2023-10-24T09-25-20.725516.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-10-24T09-25-20.725516.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_10_24T09_25_20.725516
      • path: **/details_harness|gsm8k|5_2023-10-24T09-25-20.725516.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-10-24T09-25-20.725516.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_09_13T01_25_14.196261
      • path: **/details_harness|hellaswag|10_2023-09-13T01-25-14.196261.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-09-13T01-25-14.196261.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_09_13T01_25_14.196261
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-management|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-09-13T01-25-14.196261.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-09-13T01-25-14.196261.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