five

open-llm-leaderboard/details_Henk717__chronoboros-33B

收藏
Hugging Face2023-10-12 更新2024-03-04 收录
下载链接：
https://hf-mirror.com/datasets/open-llm-leaderboard/details_Henk717__chronoboros-33B
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Henk717/chronoboros-33B在Open LLM Leaderboard上的评估运行期间自动创建的。它由64个配置组成，每个配置对应一个评估任务。数据集是从3次运行中生成的，每次运行在每个配置中表示为特定的分割。分割使用运行的时间戳命名，且“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。可以使用`datasets`库中的`load_dataset`函数加载该数据集。

This dataset was automatically created during the evaluation runs of the model Henk717/chronoboros-33B on the Open LLM Leaderboard. It consists of 64 configurations, each corresponding to one evaluation task. The dataset is generated from 3 runs, where each run is represented as a specific split under each configuration. The splits are named using the timestamp of the corresponding run, and the "train" split always points to the most recent results. An additional configuration named "results" stores the aggregated results across all runs, which are used to calculate and display the aggregate metrics on the Open LLM Leaderboard. This dataset can be loaded using the `load_dataset` function from the `datasets` library.
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 Henk717/chronoboros-33B 进行评估运行期间自动创建的，用于 Open LLM Leaderboard。

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：数据集由3次运行创建。每次运行在每个配置中都有特定的分割，分割名称使用运行的时间戳。
  • 训练分割："train" 分割始终指向最新的结果。
  • 结果配置：一个额外的配置 "results" 存储所有运行的聚合结果，用于计算和显示聚合指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Henk717__chronoboros-33B", "harness_winogrande_5", split="train")

最新结果

这些是最新结果的示例，来自2023-10-12T16:31:44.204572的运行： python { "all": { "em": 0.001572986577181208, "em_stderr": 0.00040584511324177333, "f1": 0.06618078859060408, "f1_stderr": 0.001414224388811973, "acc": 0.4767932464203287, "acc_stderr": 0.010503355727238265 }, "harness|drop|3": { "em": 0.001572986577181208, "em_stderr": 0.00040584511324177333, "f1": 0.06618078859060408, "f1_stderr": 0.001414224388811973 }, "harness|gsm8k|5": { "acc": 0.15011372251705837, "acc_stderr": 0.009838590860906965 }, "harness|winogrande|5": { "acc": 0.8034727703235991, "acc_stderr": 0.011168120593569565 } }

配置详情

以下是部分配置的详细信息：

配置：harness_arc_challenge_25

  • 分割：2023_07_31T13_28_54.015412
    • 路径：
      • **/details_harness|arc:challenge|25_2023-07-31T13:28:54.015412.parquet
  • 分割：latest
    • 路径：
      • **/details_harness|arc:challenge|25_2023-07-31T13:28:54.015412.parquet

配置：harness_drop_3

  • 分割：2023_09_22T18_14_16.970701
    • 路径：
      • **/details_harness|drop|3_2023-09-22T18-14-16.970701.parquet
  • 分割：2023_10_12T16_31_44.204572
    • 路径：
      • **/details_harness|drop|3_2023-10-12T16-31-44.204572.parquet
  • 分割：latest
    • 路径：
      • **/details_harness|drop|3_2023-10-12T16-31-44.204572.parquet

配置：harness_gsm8k_5

  • 分割：2023_09_22T18_14_16.970701
    • 路径：
      • **/details_harness|gsm8k|5_2023-09-22T18-14-16.970701.parquet
  • 分割：2023_10_12T16_31_44.204572
    • 路径：
      • **/details_harness|gsm8k|5_2023-10-12T16-31-44.204572.parquet
  • 分割：latest
    • 路径：
      • **/details_harness|gsm8k|5_2023-10-12T16-31-44.204572.parquet

配置：harness_hellaswag_10

  • 分割：2023_07_31T13_28_54.015412
    • 路径：
      • **/details_harness|hellaswag|10_2023-07-31T13:28:54.015412.parquet
  • 分割：latest
    • 路径：
      • **/details_harness|hellaswag|10_2023-07-31T13:28:54.015412.parquet

配置：harness_hendrycksTest_5

  • 分割：2023_07_31T13_28_54.015412
    • 路径：
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-management|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-07-31T13:28:54.015412.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-07-31T13:28:5
5,000+
优质数据集
54 个
任务类型
进入经典数据集
关于我们
遇见数据集——让每个数据集都被发现，让每一次遇见都有价值。
数发科官网www.sfktec.com.cn
联系我们
youjian
selectdataset@iotsh.com.cn
商业合作
数据驱动未来，携手共赢发展
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