five

open-llm-leaderboard-old/details_alnrg2arg__test3_sft_16bit

收藏
Hugging Face2024-01-27 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_alnrg2arg__test3_sft_16bit
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 alnrg2arg/test3_sft_16bit 在 Open LLM Leaderboard 上的评估运行期间自动创建的。它由 63 个配置组成,每个配置对应一个被评估的任务。数据集来源于 2 次运行,每次运行在每个配置中表示为特定的分割,分割名称使用运行的时间戳。train 分割始终指向最新的结果。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。可以使用 datasets 库中的 load_dataset 函数加载该数据集,指定数据集名称以及所需的配置和分割。

该数据集是在模型 alnrg2arg/test3_sft_16bit 在 Open LLM Leaderboard 上的评估运行期间自动创建的。它由 63 个配置组成,每个配置对应一个被评估的任务。数据集来源于 2 次运行,每次运行在每个配置中表示为特定的分割,分割名称使用运行的时间戳。train 分割始终指向最新的结果。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。可以使用 datasets 库中的 load_dataset 函数加载该数据集,指定数据集名称以及所需的配置和分割。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集基本信息

  • 数据集名称: Evaluation run of alnrg2arg/test3_sft_16bit
  • 数据集描述: 该数据集是在模型 alnrg2arg/test3_sft_16bitOpen LLM Leaderboard 上的评估运行期间自动创建的。
  • 数据集组成: 包含63个配置,每个配置对应一个评估任务。
  • 数据集来源: 从2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。
  • 额外配置: "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_alnrg2arg__test3_sft_16bit", "harness_winogrande_5", split="train")

最新结果

  • 最新结果来源: 2024-01-27T16:26:18.624159 运行
  • 结果示例: python { "all": { "acc": 0.652927958678689, "acc_stderr": 0.0321169960910649, "acc_norm": 0.6519652759500019, "acc_norm_stderr": 0.03279242565970157, "mc1": 0.576499388004896, "mc1_stderr": 0.01729742144853475, "mc2": 0.6976711663625277, "mc2_stderr": 0.015093001598591628 }, "harness|arc:challenge|25": { "acc": 0.7150170648464164, "acc_stderr": 0.013191348179838793, "acc_norm": 0.735494880546075, "acc_norm_stderr": 0.012889272949313368 }, "harness|hellaswag|10": { "acc": 0.7229635530770763, "acc_stderr": 0.004466200055292544, "acc_norm": 0.8886675960963951, "acc_norm_stderr": 0.0031390048159258633 }, ... }

配置信息

  • 配置名称: harness_arc_challenge_25

    • 数据文件:
      • 分割: 2024_01_27T16_20_56.717663
        • 路径: **/details_harness|arc:challenge|25_2024-01-27T16-20-56.717663.parquet
      • 分割: 2024_01_27T16_26_18.624159
        • 路径: **/details_harness|arc:challenge|25_2024-01-27T16-26-18.624159.parquet
      • 分割: latest
        • 路径: **/details_harness|arc:challenge|25_2024-01-27T16-26-18.624159.parquet
  • 配置名称: harness_gsm8k_5

    • 数据文件:
      • 分割: 2024_01_27T16_20_56.717663
        • 路径: **/details_harness|gsm8k|5_2024-01-27T16-20-56.717663.parquet
      • 分割: 2024_01_27T16_26_18.624159
        • 路径: **/details_harness|gsm8k|5_2024-01-27T16-26-18.624159.parquet
      • 分割: latest
        • 路径: **/details_harness|gsm8k|5_2024-01-27T16-26-18.624159.parquet
  • 配置名称: harness_hellaswag_10

    • 数据文件:
      • 分割: 2024_01_27T16_20_56.717663
        • 路径: **/details_harness|hellaswag|10_2024-01-27T16-20-56.717663.parquet
      • 分割: 2024_01_27T16_26_18.624159
        • 路径: **/details_harness|hellaswag|10_2024-01-27T16-26-18.624159.parquet
      • 分割: latest
        • 路径: **/details_harness|hellaswag|10_2024-01-27T16-26-18.624159.parquet
  • 配置名称: harness_hendrycksTest_5

    • 数据文件:
      • 分割: 2024_01_27T16_20_56.717663
        • 路径:
          • **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-anatomy|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-astronomy|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-business_ethics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-college_biology|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-college_chemistry|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-college_computer_science|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-college_mathematics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-college_medicine|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-college_physics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-computer_security|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-econometrics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-formal_logic|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-global_facts|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_biology|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_computer_science|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_european_history|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_geography|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_mathematics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_microeconomics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_physics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_psychology|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_statistics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_us_history|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-high_school_world_history|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-human_aging|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-human_sexuality|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-international_law|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-jurisprudence|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-logical_fallacies|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-machine_learning|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-management|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-marketing|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-medical_genetics|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-miscellaneous|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-moral_disputes|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-moral_scenarios|5_2024-01-27T16-20-56.717663.parquet
          • **/details_harness|hendrycksTest-nutrition|5_2024-01-27T16-20-56
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