five

open-llm-leaderboard-old/details_huseyinatahaninan__phi-2-dpo

收藏
Hugging Face2024-02-12 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_huseyinatahaninan__phi-2-dpo
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huseyinatahaninan/phi-2-dpo的评估运行期间自动创建的,用于Open LLM Leaderboard的评估。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分割始终指向最新的结果。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。

该数据集是在模型huseyinatahaninan/phi-2-dpo的评估运行期间自动创建的,用于Open LLM Leaderboard的评估。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分割始终指向最新的结果。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-old
原始信息汇总

数据集概述

该数据集是在评估模型huseyinatahaninan/phi-2-dpoOpen LLM Leaderboard上的自动创建的。数据集包含63个配置,每个配置对应一个评估任务。数据集从1次运行中创建,每个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。

数据集结构

数据集包含以下配置:

  • config_name: harness_arc_challenge_25

    • data_files:
      • split: 2024_02_12T21_58_15.192256
        • path: **/details_harness|arc:challenge|25_2024-02-12T21-58-15.192256.parquet
      • split: latest
        • path: **/details_harness|arc:challenge|25_2024-02-12T21-58-15.192256.parquet
  • config_name: harness_gsm8k_5

    • data_files:
      • split: 2024_02_12T21_58_15.192256
        • path: **/details_harness|gsm8k|5_2024-02-12T21-58-15.192256.parquet
      • split: latest
        • path: **/details_harness|gsm8k|5_2024-02-12T21-58-15.192256.parquet
  • config_name: harness_hellaswag_10

    • data_files:
      • split: 2024_02_12T21_58_15.192256
        • path: **/details_harness|hellaswag|10_2024-02-12T21-58-15.192256.parquet
      • split: latest
        • path: **/details_harness|hellaswag|10_2024-02-12T21-58-15.192256.parquet
  • config_name: harness_hendrycksTest_5

    • data_files:
      • split: 2024_02_12T21_58_15.192256
        • path:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-management|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-philosophy|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-prehistory|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-professional_accounting|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-professional_law|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-professional_medicine|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-professional_psychology|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-public_relations|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-security_studies|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-sociology|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-us_foreign_policy|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-virology|5_2024-02-12T21-58-15.192256.parquet
          • **/details_harness|hendrycksTest-world_religions|5_2024-02-12T21-58-15.192256.parquet

最新结果

以下是2024-02-12T21:58:15.192256运行的最新结果:

python { "all": { "acc": 0.5870761708653485, "acc_stderr": 0.03369469581974977, "acc_norm": 0.5884353168964569, "acc_norm_stderr": 0.034381836157511524, "mc1": 0.3157894736842105, "mc1_stderr": 0.016272287957916912, "mc2": 0.45354154186159823, "mc2_stderr": 0.015221463708711597 }, "harness|arc:challenge|25": { "acc": 0.6040955631399317, "acc_stderr": 0.014291228393536588, "acc_norm": 0.6

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