five

open-llm-leaderboard-old/details_flemmingmiguel__MBX-7B-v2

收藏
Hugging Face2024-01-29 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_flemmingmiguel__MBX-7B-v2
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型flemmingmiguel/MBX-7B-v2时自动创建的,包含63个配置,每个配置对应一个评估任务。数据集由6次运行生成,每次运行的结果作为一个特定的分割存储,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。

该数据集是在评估模型flemmingmiguel/MBX-7B-v2时自动创建的,包含63个配置,每个配置对应一个评估任务。数据集由6次运行生成,每次运行的结果作为一个特定的分割存储,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。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 flemmingmiguel/MBX-7B-v2Open LLM Leaderboard 上的运行过程中自动创建的。数据集包含 63 个配置,每个配置对应一个评估任务。

数据集结构

数据集由 6 次运行创建,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。

数据集加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_flemmingmiguel__MBX-7B-v2", "harness_winogrande_5", split="train")

最新结果

以下是 2024-01-29T01:58:59.417725 运行 的最新结果:

python { "all": { "acc": 0.6539700518752208, "acc_stderr": 0.03202984813423276, "acc_norm": 0.6533129460161269, "acc_norm_stderr": 0.0326981687674504, "mc1": 0.5679314565483476, "mc1_stderr": 0.01734120239498833, "mc2": 0.7021318083151993, "mc2_stderr": 0.014958406447536246 }, "harness|arc:challenge|25": { "acc": 0.7056313993174061, "acc_stderr": 0.01331852846053942, "acc_norm": 0.735494880546075, "acc_norm_stderr": 0.012889272949313368 }, "harness|hellaswag|10": { "acc": 0.717486556462856, "acc_stderr": 0.004493015945599716, "acc_norm": 0.8849830711013742, "acc_norm_stderr": 0.0031839033919416975 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.33, "acc_stderr": 0.04725815626252605, "acc_norm": 0.33, "acc_norm_stderr": 0.04725815626252605 }, # 其他任务的结果省略... }

配置详情

数据集包含多个配置,每个配置对应不同的任务和运行结果。以下是部分配置的示例:

  • config_name: harness_arc_challenge_25

    • 数据文件分割和路径:
      • split: 2024_01_23T07_44_55.248174
        • path: **/details_harness|arc:challenge|25_2024-01-23T07-44-55.248174.parquet
      • split: 2024_01_23T18_17_28.506209
        • path: **/details_harness|arc:challenge|25_2024-01-23T18-17-28.506209.parquet
      • split: 2024_01_23T23_42_41.090832
        • path: **/details_harness|arc:challenge|25_2024-01-23T23-42-41.090832.parquet
      • split: 2024_01_25T06_20_52.562143
        • path: **/details_harness|arc:challenge|25_2024-01-25T06-20-52.562143.parquet
      • split: 2024_01_25T22_47_13.183013
        • path: **/details_harness|arc:challenge|25_2024-01-25T22-47-13.183013.parquet
      • split: 2024_01_29T01_58_59.417725
        • path: **/details_harness|arc:challenge|25_2024-01-29T01-58-59.417725.parquet
      • split: latest
        • path: **/details_harness|arc:challenge|25_2024-01-29T01-58-59.417725.parquet
  • config_name: harness_gsm8k_5

    • 数据文件分割和路径:
      • split: 2024_01_23T07_44_55.248174
        • path: **/details_harness|gsm8k|5_2024-01-23T07-44-55.248174.parquet
      • split: 2024_01_23T18_17_28.506209
        • path: **/details_harness|gsm8k|5_2024-01-23T18-17-28.506209.parquet
      • split: 2024_01_23T23_42_41.090832
        • path: **/details_harness|gsm8k|5_2024-01-23T23-42-41.090832.parquet
      • split: 2024_01_25T06_20_52.562143
        • path: **/details_harness|gsm8k|5_2024-01-25T06-20-52.562143.parquet
      • split: 2024_01_25T22_47_13.183013
        • path: **/details_harness|gsm8k|5_2024-01-25T22-47-13.183013.parquet
      • split: 2024_01_29T01_58_59.417725
        • path: **/details_harness|gsm8k|5_2024-01-29T01-58-59.417725.parquet
      • split: latest
        • path: **/details_harness|gsm8k|5_2024-01-29T01-58-59.417725.parquet
  • config_name: harness_hellaswag_10

    • 数据文件分割和路径:
      • split: 2024_01_23T07_44_55.248174
        • path: **/details_harness|hellaswag|10_2024-01-23T07-44-55.248174.parquet
      • split: 2024_01_23T18_17_28.506209
        • path: **/details_harness|hellaswag|10_2024-01-23T18-17-28.506209.parquet
      • split: 2024_01_23T23_42_41.090832
        • path: **/details_harness|hellaswag|10_2024-01-23T23-42-41.090832.parquet
      • split: 2024_01_25T06_20_52.562143
        • path: **/details_harness|hellaswag|10_2024-01-25T06-20-52.562143.parquet
      • split: 2024_01_25T22_47_13.183013
        • path: **/details_harness|hellaswag|10_2024-01-25T22-47-13.183013.parquet
      • split: 2024_01_29T01_58_59.417725
        • path: **/details_harness|hellaswag|10_2024-01-29T01-58-59.417725.parquet
      • split: latest
        • path: **/details_harness|hellaswag|10_2024-01-29T01-58-59.417725.parquet

数据集配置

数据集包含多个配置,每个配置对应不同的任务和运行结果。以下是部分配置的示例:

  • config_name: harness_arc_challenge_25

    • 数据文件分割和路径:
      • split: 2024_01_23T07_44_55.248174
        • path: **/details_harness|arc:challenge|25_2024-01-23T07-44-55.248174.parquet
      • split: 2024_01_23T18_17_28.506209
        • path: **/details_harness|arc:challenge|25_2024-01-23T18-17-28.506209.parquet
      • split: 2024_01_23T23_42_41.090832
        • path: **/details_harness|arc:challenge|25_2024-01-23T23-42-41.090832.parquet
      • split: 2024_01_25T06_20_52.562143
        • path: **/details_harness|arc:challenge|25_2024-01-25T06-20-52.562143.parquet
      • split: 2024_01_25T22_47_13.183013
        • path: **/details_harness|arc:challenge|25_2024-01-25T22-47-13.183013.parquet
      • split: 2024_01_29T01_58_59.417725
        • path: **/details_harness|arc:challenge|25_2024-01-29T01-58-59.417725.parquet
      • split: latest
        • path: **/details_harness|arc:challenge|25_2024-01-29T01-58-59.417725.parquet
  • config_name: harness_gsm8k_5

    • 数据文件分割和路径:
      • split: 2024_01_23T07_44_55.248174
        • path: **/details_harness|gsm8k|5_2024-01-23T07-44-55.248174.parquet
      • split: 2024_01_23T18_17_28.506209
        • path: **/details_harness|gsm8k|5_2024-01-23T18-17-28.506209.parquet
      • split: 2024_01_23T23_42_41.090832
        • path: **/details_harness|gsm8k|5_2024-01-23T23-42-41.090832.parquet
      • split: 2024_01_25T06_20_52.562143
        • path: **/details_harness|gsm8k|5_2024-01-25T06-20-52.562143.parquet
      • split: 2024_01_25T22_47_13.183013
        • path: **/details_harness|gsm8k|5_2024-01-25T22-47-13.183013.parquet
      • split: 2024_01_29T01_58_59.417725
        • path: **/details_harness|gsm8k|5_2024-01-29T01-58-59.417725.parquet
      • split: latest
        • path: **/details_harness|gsm8k|5_2024-01-29T01-58-59.417725.parquet
  • config_name: harness_hellaswag_10

    • 数据文件分割和路径:
      • split: 2024_01_23T07_44_55.248174
        • path: **/details_harness|hellaswag|10_2024-01-23T07-44-55.248174.parquet
      • split: 2024_01_23T18_17_28.506209
        • path: **/details_harness|hellaswag|10_2024-01-23T18-17-28.506209.parquet
      • split: 2024_01_23T23_42_41.090832
        • path: **/details_harness|hellaswag|10
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