five

open-llm-leaderboard/details_lvkaokao__llama2-7b-hf-instruction-lora

收藏
Hugging Face2023-09-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_lvkaokao__llama2-7b-hf-instruction-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[lvkaokao/llama2-7b-hf-instruction-lora]时自动创建的,评估过程在[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)上进行。数据集由64个配置组成,每个配置对应一个评估任务。数据集由2次运行生成,每次运行的结果作为特定配置中的一个分割,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,还有一个名为results的配置,存储了所有运行的聚合结果,用于计算和显示[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/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

该数据集是在评估模型lvkaokao/llama2-7b-hf-instruction-loraOpen LLM Leaderboard上的运行过程中自动创建的。

数据集组成

  • 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。
  • 一个额外的配置"results"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_lvkaokao__llama2-7b-hf-instruction-lora", "harness_winogrande_5", split="train")

最新结果

以下是2023-09-22T19:01:17.427442运行的最新结果: python { "all": { "em": 0.32466442953020136, "em_stderr": 0.004795311441664345, "f1": 0.372114093959732, "f1_stderr": 0.00471907408435466, "acc": 0.4202347692309533, "acc_stderr": 0.010254299592459359 }, "harness|drop|3": { "em": 0.32466442953020136, "em_stderr": 0.004795311441664345, "f1": 0.372114093959732, "f1_stderr": 0.00471907408435466 }, "harness|gsm8k|5": { "acc": 0.09855951478392722, "acc_stderr": 0.008210320350946328 }, "harness|winogrande|5": { "acc": 0.7419100236779794, "acc_stderr": 0.01229827883397239 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|arc:challenge|25_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-08-09T14:42:44.392764.parquet
  • harness_drop_3

    • 分割:2023_09_22T19_01_17.427442
    • 路径:**/details_harness|drop|3_2023-09-22T19-01-17.427442.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-09-22T19-01-17.427442.parquet
  • harness_gsm8k_5

    • 分割:2023_09_22T19_01_17.427442
    • 路径:**/details_harness|gsm8k|5_2023-09-22T19-01-17.427442.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-09-22T19-01-17.427442.parquet
  • harness_hellaswag_10

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hellaswag|10_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-09T14:42:44.392764.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_08_09T14_42_44.392764
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T14:42:44.392764.parquet
    • 分割:latest
    • 路径:`**/details_harness|hendrycksTest-electrical_engineering|5_2
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