five

open-llm-leaderboard/details_KnutJaegersberg__deacon-3b

收藏
Hugging Face2023-10-28 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_KnutJaegersberg__deacon-3b
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KnutJaegersberg/deacon-3b的评估运行期间自动创建的,用于在Open LLM Leaderboard上进行评估。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,每次运行都可以在特定配置的分割中找到,分割以运行的时间戳命名。train分割始终指向最新的结果。此外,results配置存储了所有运行的聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。

This dataset was automatically generated during the evaluation runs of the model KnutJaegersberg/deacon-3b, for the purpose of evaluation on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to a single evaluation task. The dataset is created from two runs, with each run accessible via the split of its corresponding specific configuration, where the split is named after the timestamp of the run. The train split always points to the most up-to-date results. Additionally, the "results" configuration stores the aggregated results across all runs, and is utilized to compute 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

数据集简介

该数据集是在评估模型 KnutJaegersberg/deacon-3bOpen LLM Leaderboard 上的自动创建的。

数据集结构

  • 配置数量:64个配置,每个配置对应一个评估任务。
  • 运行次数:数据集从2次运行中创建。每个运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。
  • 训练分片:"train" 分片总是指向最新的结果。
  • 结果配置: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_KnutJaegersberg__deacon-3b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-28T13:23:04.115502 运行的最新结果: python { "all": { "em": 0.0009437919463087249, "em_stderr": 0.0003144653119413187, "f1": 0.05062919463087265, "f1_stderr": 0.0012970020903289405, "acc": 0.32509979517380905, "acc_stderr": 0.007564621001375068 }, "harness|drop|3": { "em": 0.0009437919463087249, "em_stderr": 0.0003144653119413187, "f1": 0.05062919463087265, "f1_stderr": 0.0012970020903289405 }, "harness|gsm8k|5": { "acc": 0.0037907505686125853, "acc_stderr": 0.0016927007401501847 }, "harness|winogrande|5": { "acc": 0.6464088397790055, "acc_stderr": 0.013436541262599952 } }

配置详情

以下是数据集的配置详情:

配置列表

  • harness_arc_challenge_25

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|arc:challenge|25_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|arc:challenge|25_2023-09-18T14-47-42.541004.parquet
  • harness_drop_3

    • 分片:2023_10_28T13_23_04.115502
    • 路径:**/details_harness|drop|3_2023-10-28T13-23-04.115502.parquet
    • 分片:latest
    • 路径:**/details_harness|drop|3_2023-10-28T13-23-04.115502.parquet
  • harness_gsm8k_5

    • 分片:2023_10_28T13_23_04.115502
    • 路径:**/details_harness|gsm8k|5_2023-10-28T13-23-04.115502.parquet
    • 分片:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-28T13-23-04.115502.parquet
  • harness_hellaswag_10

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hellaswag|10_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hellaswag|10_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_anatomy_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_astronomy_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_college_biology_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_college_physics_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_computer_security_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_econometrics_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-18T14-47-42.541004.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-18T14-47-42.541004.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分片:2023_09_18T14_47_42.541004
    • 路径:`**/details_harness|hendry
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