five

open-llm-leaderboard-old/details_ToolBench__ToolLLaMA-7b-LoRA

收藏
Hugging Face2023-09-18 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_ToolBench__ToolLLaMA-7b-LoRA
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在评估模型ToolBench/ToolLLaMA-7b-LoRA时自动生成的,用于Open LLM Leaderboard的评估。数据集包含64个配置,每个配置对应一个评估任务。数据集由2次运行生成,每次运行的结果存储在不同的分割中,分割名称使用运行的时间戳命名。train分割始终指向最新的结果。此外,数据集还包含一个名为results的配置,用于存储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。

该数据集是在评估模型ToolBench/ToolLLaMA-7b-LoRA时自动生成的,用于Open LLM Leaderboard的评估。数据集包含64个配置,每个配置对应一个评估任务。数据集由2次运行生成,每次运行的结果存储在不同的分割中,分割名称使用运行的时间戳命名。train分割始终指向最新的结果。此外,数据集还包含一个名为results的配置,用于存储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

数据集摘要

该数据集是在对模型 ToolBench/ToolLLaMA-7b-LoRA 进行评估运行期间自动创建的。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2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。

最新结果

以下是 2023-09-18T01:38:32.661486 运行 的最新结果:

python { "all": { "em": 0.0010486577181208054, "em_stderr": 0.00033145814652191404, "f1": 0.056097944630872455, "f1_stderr": 0.001312187728090684, "acc": 0.40586103293913917, "acc_stderr": 0.00960950347641371 }, "harness|drop|3": { "em": 0.0010486577181208054, "em_stderr": 0.00033145814652191404, "f1": 0.056097944630872455, "f1_stderr": 0.001312187728090684 }, "harness|gsm8k|5": { "acc": 0.06823351023502654, "acc_stderr": 0.006945358944067431 }, "harness|winogrande|5": { "acc": 0.7434885556432518, "acc_stderr": 0.01227364800875999 } }

配置详情

数据集包含以下配置及其对应的数据文件路径:

  • harness_arc_challenge_25

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|arc:challenge|25_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2023-08-09T17:09:39.923597.parquet
  • harness_drop_3

    • 分割: 2023_09_18T01_38_32.661486
      • 路径: **/details_harness|drop|3_2023-09-18T01-38-32.661486.parquet
    • 分割: latest
      • 路径: **/details_harness|drop|3_2023-09-18T01-38-32.661486.parquet
  • harness_gsm8k_5

    • 分割: 2023_09_18T01_38_32.661486
      • 路径: **/details_harness|gsm8k|5_2023-09-18T01-38-32.661486.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2023-09-18T01-38-32.661486.parquet
  • harness_hellaswag_10

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hellaswag|10_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T17:09:39.923597.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-09T17:09:39.923597.parquet
        • ...(其他路径省略)
    • 分割: latest
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T17:09:39.923597.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-09T17:09:39.923597.parquet
        • ...(其他路径省略)
  • harness_hendrycksTest_abstract_algebra_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-08-09T17:09:39.923597.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割: 2023_08_09T17_09_39.923597
      • 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T17:09:39.923597.parquet
    • 分割: latest
      • 路径: `**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T17:
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