five

open-llm-leaderboard-old/details_kittn__mistral-7B-v0.1-hf

收藏
Hugging Face2023-10-25 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_kittn__mistral-7B-v0.1-hf
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上对模型kittn/mistral-7B-v0.1-hf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4次运行创建,每次运行的结果作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,用于计算和显示Open LLM Leaderboard上的聚合指标。

该数据集是在Open LLM Leaderboard上对模型kittn/mistral-7B-v0.1-hf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4次运行创建,每次运行的结果作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,用于计算和显示Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard-old
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 kittn/mistral-7B-v0.1-hf 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:数据集包含4次运行结果。每次运行的结果作为一个特定的分割(split)存储,分割名称使用运行的时间戳。
  • 最新结果:"train" 分割始终指向最新的结果。
  • 结果汇总:一个额外的配置 "results" 存储所有运行的汇总结果,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_kittn__mistral-7B-v0.1-hf", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-25T07:14:20.510408 运行的最新结果: python { "all": { "em": 0.0014681208053691276, "em_stderr": 0.00039210421902984423, "f1": 0.061358011744966516, "f1_stderr": 0.0013673296029707471, "acc": 0.48246949896868846, "acc_stderr": 0.011090221236468356 }, "harness|drop|3": { "em": 0.0014681208053691276, "em_stderr": 0.00039210421902984423, "f1": 0.061358011744966516, "f1_stderr": 0.0013673296029707471 }, "harness|gsm8k|5": { "acc": 0.18119787717968158, "acc_stderr": 0.010609827611527364 }, "harness|winogrande|5": { "acc": 0.7837411207576953, "acc_stderr": 0.01157061486140935 } }

配置详情

以下是部分配置及其数据文件路径:

  • harness_arc_challenge_25

    • 分割:2023_10_03T19_42_22.443456
      • 路径:**/details_harness|arc:challenge|25_2023-10-03T19-42-22.443456.parquet
    • 分割:2023_10_03T19_50_01.602909
      • 路径:**/details_harness|arc:challenge|25_2023-10-03T19-50-01.602909.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-10-03T19-50-01.602909.parquet
  • harness_drop_3

    • 分割:2023_10_25T02_00_59.086242
      • 路径:**/details_harness|drop|3_2023-10-25T02-00-59.086242.parquet
    • 分割:2023_10_25T07_14_20.510408
      • 路径:**/details_harness|drop|3_2023-10-25T07-14-20.510408.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-25T07-14-20.510408.parquet
  • harness_gsm8k_5

    • 分割:2023_10_25T02_00_59.086242
      • 路径:**/details_harness|gsm8k|5_2023-10-25T02-00-59.086242.parquet
    • 分割:2023_10_25T07_14_20.510408
      • 路径:**/details_harness|gsm8k|5_2023-10-25T07-14-20.510408.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-25T07-14-20.510408.parquet
  • harness_hellaswag_10

    • 分割:2023_10_03T19_42_22.443456
      • 路径:**/details_harness|hellaswag|10_2023-10-03T19-42-22.443456.parquet
    • 分割:2023_10_03T19_50_01.602909
      • 路径:**/details_harness|hellaswag|10_2023-10-03T19-50-01.602909.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-10-03T19-50-01.602909.parquet
  • harness_hendrycksTest_5

    • 分割:2023_10_03T19_42_22.443456
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-formal_logic|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-global_facts|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_biology|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_chemistry|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_computer_science|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_european_history|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_geography|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_mathematics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_physics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_psychology|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_statistics|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_us_history|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_world_history|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-human_aging|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-human_sexuality|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-international_law|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-jurisprudence|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-logical_fallacies|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-machine_learning|5_2023-10-03T19-42-22.443456.parquet
      • 路径:**/details_harness|hendrycksTest-management|5_2023-10-03T19-42-22.443456.parquet
      • 路径:`**/details_harness|hendrycksTest-marketing|5_2023-10-03T19-42-22.443456.
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