five

open-llm-leaderboard/details_wei123602__llama-13b-FINETUNE3

收藏
Hugging Face2023-10-25 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_wei123602__llama-13b-FINETUNE3
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wei123602/llama-13b-FINETUNE3评估运行期间自动创建的,用于Open LLM排行榜。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次运行创建,每次运行作为一个特定的时间戳命名的分割。train分割始终指向最新结果。额外的results配置存储了运行的所有聚合结果,用于在排行榜上计算和显示聚合指标。数据集结构化,具有针对不同任务的特定配置,每个配置都有按时间戳和latest分割的数据文件。

该数据集是在模型wei123602/llama-13b-FINETUNE3评估运行期间自动创建的,用于Open LLM排行榜。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次运行创建,每次运行作为一个特定的时间戳命名的分割。train分割始终指向最新结果。额外的results配置存储了运行的所有聚合结果,用于在排行榜上计算和显示聚合指标。数据集结构化,具有针对不同任务的特定配置,每个配置都有按时间戳和latest分割的数据文件。
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 wei123602/llama-13b-FINETUNE3Open LLM Leaderboard 上的自动创建的。

数据集组成

  • 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 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_wei123602__llama-13b-FINETUNE3", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-25T20:56:48.132337 运行的最新结果: python { "all": { "em": 0.10360738255033557, "em_stderr": 0.003120930790921416, "f1": 0.14798552852348912, "f1_stderr": 0.003214007613815376, "acc": 0.4442352766589695, "acc_stderr": 0.010435544785566055 }, "harness|drop|3": { "em": 0.10360738255033557, "em_stderr": 0.003120930790921416, "f1": 0.14798552852348912, "f1_stderr": 0.003214007613815376 }, "harness|gsm8k|5": { "acc": 0.12130401819560273, "acc_stderr": 0.00899288849727557 }, "harness|winogrande|5": { "acc": 0.7671665351223362, "acc_stderr": 0.01187820107385654 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|arc:challenge|25_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-09-13T02-24-38.254919.parquet
  • harness_drop_3

    • 分割:2023_10_25T20_56_48.132337
    • 路径:**/details_harness|drop|3_2023-10-25T20-56-48.132337.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-10-25T20-56-48.132337.parquet
  • harness_gsm8k_5

    • 分割:2023_10_25T20_56_48.132337
    • 路径:**/details_harness|gsm8k|5_2023-10-25T20-56-48.132337.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-25T20-56-48.132337.parquet
  • harness_hellaswag_10

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hellaswag|10_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T02-24-38.254919.parquet 等 50 个文件
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T02-24-38.254919.parquet 等 50 个文件
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-13T02-24-38.254919.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_09_13T02_24_38.254919
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-13T02-24-38.254919.parquet
    • 分割:latest
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