five

open-llm-leaderboard-old/details_uukuguy__zephyr-7b-alpha-dare-0.85

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_uukuguy__zephyr-7b-alpha-dare-0.85
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 uukuguy/zephyr-7b-alpha-dare-0.85 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 63 个配置组成,每个配置对应一个被评估的任务。它包含一次运行的结果,每次运行在每个配置中表示为特定的分割。train 分割始终指向最新结果。一个名为 results 的额外配置存储了运行的所有聚合结果,这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个使用 Python 中的 datasets 库加载运行详情的示例。

该数据集是在模型 uukuguy/zephyr-7b-alpha-dare-0.85 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 63 个配置组成,每个配置对应一个被评估的任务。它包含一次运行的结果,每次运行在每个配置中表示为特定的分割。train 分割始终指向最新结果。一个名为 results 的额外配置存储了运行的所有聚合结果,这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个使用 Python 中的 datasets 库加载运行详情的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 uukuguy/zephyr-7b-alpha-dare-0.85Open LLM Leaderboard 上的自动创建的。

数据集组成

  • 数据集包含 63 个配置,每个配置对应一个评估任务。
  • 数据集从 1 次运行中创建,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" 分割始终指向最新结果。
  • 额外的 "results" 配置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

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_uukuguy__zephyr-7b-alpha-dare-0.85", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果 的摘要:

python { "all": { "acc": 0.6405125012890543, "acc_stderr": 0.0322440782989453, "acc_norm": 0.6457442431541438, "acc_norm_stderr": 0.032888705588954556, "mc1": 0.29498164014687883, "mc1_stderr": 0.015964400965589657, "mc2": 0.4441404853042373, "mc2_stderr": 0.014450558004670922 }, "harness|arc:challenge|25": { "acc": 0.5767918088737202, "acc_stderr": 0.01443803622084803, "acc_norm": 0.6117747440273038, "acc_norm_stderr": 0.01424161420741405 }, "harness|hellaswag|10": { "acc": 0.6387173869747063, "acc_stderr": 0.004793904922401889, "acc_norm": 0.8366859191396137, "acc_norm_stderr": 0.0036889652317335197 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.04793724854411021, "acc_norm": 0.35, "acc_norm_stderr": 0.04793724854411021 }, # 其他任务的结果... }

配置详情

  • harness_arc_challenge_25

    • 分割: 2023_12_04T16_03_30.985884
    • 路径: **/details_harness|arc:challenge|25_2023-12-04T16-03-30.985884.parquet
    • 分割: latest
    • 路径: **/details_harness|arc:challenge|25_2023-12-04T16-03-30.985884.parquet
  • harness_gsm8k_5

    • 分割: 2023_12_04T16_03_30.985884
    • 路径: **/details_harness|gsm8k|5_2023-12-04T16-03-30.985884.parquet
    • 分割: latest
    • 路径: **/details_harness|gsm8k|5_2023-12-04T16-03-30.985884.parquet
  • harness_hellaswag_10

    • 分割: 2023_12_04T16_03_30.985884
    • 路径: **/details_harness|hellaswag|10_2023-12-04T16-03-30.985884.parquet
    • 分割: latest
    • 路径: **/details_harness|hellaswag|10_2023-12-04T16-03-30.985884.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_12_04T16_03_30.985884
    • 路径:
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-management|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-12-04T16-03-30.985884.parquet
      • **/details_harness|hendrycksTest-professional_law|5_2023-12-04T16-03-30.985884.parquet
二维码
社区交流群
二维码
科研交流群
商业服务