five

open-llm-leaderboard-old/details_bigscience__bloom-7b1

收藏
Hugging Face2023-12-04 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_bigscience__bloom-7b1
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 bigscience/bloom-7b1 在 Open LLM Leaderboard 上的评估运行期间自动生成的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。数据集是从 10 次运行中创建的,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。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。README 还提供了如何使用 Python 代码加载运行中的详细信息的示例。

该数据集是在模型 bigscience/bloom-7b1 在 Open LLM Leaderboard 上的评估运行期间自动生成的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。数据集是从 10 次运行中创建的,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。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。README 还提供了如何使用 Python 代码加载运行中的详细信息的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集简介

数据集名为 "Evaluation run of bigscience/bloom-7b1",是在评估模型 bigscience/bloom-7b1Open LLM Leaderboard 上的自动创建的。

数据集结构

  • 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 10 次运行中创建,每次运行在每个配置中作为一个特定的分割存在,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

以下是加载特定运行详细信息的示例代码: python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_bigscience__bloom-7b1", "harness_gsm8k_5", split="train")

最新结果

以下是 2023-12-04T13:10:02.911977 运行的最新结果: python { "all": { "acc": 0.013646702047005308, "acc_stderr": 0.0031957470754807806 }, "harness|gsm8k|5": { "acc": 0.013646702047005308, "acc_stderr": 0.0031957470754807806 } }

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_08_01T14_42_42.953249
    • 路径: **/details_harness|arc:challenge|25_2023-08-01T14:42:42.953249.parquet
  • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2023-08-01T14:42:42.953249.parquet

harness_drop_3

  • 分割: 2023_10_17T22_39_12.950006
    • 路径: **/details_harness|drop|3_2023-10-17T22-39-12.950006.parquet
  • 分割: 2023_10_19T04_28_54.166367
    • 路径: **/details_harness|drop|3_2023-10-19T04-28-54.166367.parquet
  • 分割: latest
    • 路径: **/details_harness|drop|3_2023-10-19T04-28-54.166367.parquet

harness_gsm8k_5

  • 分割: 2023_10_17T22_39_12.950006
    • 路径: **/details_harness|gsm8k|5_2023-10-17T22-39-12.950006.parquet
  • 分割: 2023_10_19T04_28_54.166367
    • 路径: **/details_harness|gsm8k|5_2023-10-19T04-28-54.166367.parquet
  • 分割: 2023_12_03T15_05_35.973526
    • 路径: **/details_harness|gsm8k|5_2023-12-03T15-05-35.973526.parquet
  • 分割: 2023_12_03T16_09_00.043447
    • 路径: **/details_harness|gsm8k|5_2023-12-03T16-09-00.043447.parquet
  • 分割: 2023_12_03T16_09_29.945299
    • 路径: **/details_harness|gsm8k|5_2023-12-03T16-09-29.945299.parquet
  • 分割: 2023_12_04T09_58_57.284123
    • 路径: **/details_harness|gsm8k|5_2023-12-04T09-58-57.284123.parquet
  • 分割: 2023_12_04T09_59_23.998430
    • 路径: **/details_harness|gsm8k|5_2023-12-04T09-59-23.998430.parquet
  • 分割: 2023_12_04T13_09_04.314120
    • 路径: **/details_harness|gsm8k|5_2023-12-04T13-09-04.314120.parquet
  • 分割: 2023_12_04T13_10_02.911977
    • 路径: **/details_harness|gsm8k|5_2023-12-04T13-10-02.911977.parquet
  • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2023-12-04T13-10-02.911977.parquet

harness_hellaswag_10

  • 分割: 2023_08_01T14_42_42.953249
    • 路径: **/details_harness|hellaswag|10_2023-08-01T14:42:42.953249.parquet
  • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2023-08-01T14:42:42.953249.parquet

harness_hendrycksTest_5

  • 分割: 2023_08_01T14_42_42.953249
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-management|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-08-01T14:42:42.953249.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-01T14:42:42.953
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