five

open-llm-leaderboard/details_tricktreat__Llama-2-7b-chat-hf-guanaco-lora

收藏
Hugging Face2024-04-16 更新2024-06-12 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_tricktreat__Llama-2-7b-chat-hf-guanaco-lora
下载链接
链接失效反馈
官方服务:
资源简介:
数据集是在评估模型tricktreat/Llama-2-7b-chat-hf-guanaco-lora在Open LLM Leaderboard上的运行过程中自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集由1次运行创建,每次运行可以在每个配置中找到,运行的时间戳作为分割名称。train分割始终指向最新结果。此外,results配置存储了所有运行的聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。

数据集是在评估模型tricktreat/Llama-2-7b-chat-hf-guanaco-lora在Open LLM Leaderboard上的运行过程中自动创建的。数据集由63个配置组成,每个配置对应一个评估任务。数据集由1次运行创建,每次运行可以在每个配置中找到,运行的时间戳作为分割名称。train分割始终指向最新结果。此外,results配置存储了所有运行的聚合结果,并用于计算和显示Open LLM Leaderboard上的聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集创建

  • 创建背景:该数据集是在对模型 tricktreat/Llama-2-7b-chat-hf-guanaco-lora 进行评估运行期间自动创建的。
  • 评估平台:数据集的创建基于 Open LLM Leaderboard

数据集结构

  • 配置数量:数据集包含 63 个配置,每个配置对应一个评估任务。
  • 运行次数:数据集来自 1 次运行,每个运行结果作为一个特定的分割存在于每个配置中,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。

额外配置

  • 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_tricktreat__Llama-2-7b-chat-hf-guanaco-lora", "harness_winogrande_5", split="train")

最新结果

  • 最新结果时间戳:2024-04-16T04:25:33.258819
  • 结果示例: python { "all": { "acc": 0.4893443209680558, "acc_stderr": 0.034417544339580504, "acc_norm": 0.4944808333548298, "acc_norm_stderr": 0.0351790056723058, "mc1": 0.2839657282741738, "mc1_stderr": 0.015785370858396718, "mc2": 0.4382229323743104, "mc2_stderr": 0.014699763853417739 }, "harness|arc:challenge|25": { "acc": 0.4778156996587031, "acc_stderr": 0.014597001927076138, "acc_norm": 0.5264505119453925, "acc_norm_stderr": 0.01459093135812017 }, "harness|hellaswag|10": { "acc": 0.5690101573391755, "acc_stderr": 0.004942026200279591, "acc_norm": 0.7667795259908384, "acc_norm_stderr": 0.004220171569273325 },

    其他任务结果省略

}

配置详情

  • 配置名称harness_arc_challenge_25

    • 数据文件
      • 分割:2024_04_16T04_25_33.258819
        • 路径:**/details_harness|arc:challenge|25_2024-04-16T04-25-33.258819.parquet
      • 分割:latest
        • 路径:**/details_harness|arc:challenge|25_2024-04-16T04-25-33.258819.parquet
  • 配置名称harness_gsm8k_5

    • 数据文件
      • 分割:2024_04_16T04_25_33.258819
        • 路径:**/details_harness|gsm8k|5_2024-04-16T04-25-33.258819.parquet
      • 分割:latest
        • 路径:**/details_harness|gsm8k|5_2024-04-16T04-25-33.258819.parquet
  • 配置名称harness_hellaswag_10

    • 数据文件
      • 分割:2024_04_16T04_25_33.258819
        • 路径:**/details_harness|hellaswag|10_2024-04-16T04-25-33.258819.parquet
      • 分割:latest
        • 路径:**/details_harness|hellaswag|10_2024-04-16T04-25-33.258819.parquet
  • 配置名称harness_hendrycksTest_5

    • 数据文件
      • 分割:2024_04_16T04_25_33.258819
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-management|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2024-04-16T04-25-33.258819.parquet
          • **/details_harness|hendrycksTest-professional_medicine|5_2024-04-16T04-25-33.258819.parquet
          • `**/details_harness|hendrycksTest-professional_psychology|5_2024-04-16T04-
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