five

open-llm-leaderboard/details_RWKV__rwkv-4-7b-pile

收藏
Hugging Face2023-10-16 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_RWKV__rwkv-4-7b-pile
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型RWKV/rwkv-4-7b-pile进行评估时自动创建的。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次运行创建,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。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上的聚合指标。README还提供了一个使用`datasets`库中的`load_dataset`函数加载运行细节的示例。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集名称

  • Evaluation run of RWKV/rwkv-4-7b-pile

数据集组成

  • 该数据集包含64个配置,每个配置对应一个评估任务。
  • 数据集由2次运行结果组成,每次运行结果作为特定分片存储在每个配置中,分片名称使用运行的时间戳。
  • "train"分片始终指向最新的结果。
  • 一个额外的配置"results"存储所有运行结果的聚合数据,用于计算和显示在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_RWKV__rwkv-4-7b-pile", "harness_winogrande_5", split="train")

最新结果

  • 最新结果来自2023-10-16T04:13:51.306808的运行,包含以下指标: python { "all": { "em": 0.003145973154362416, "em_stderr": 0.0005734993648436389, "f1": 0.04605390100671147, "f1_stderr": 0.0012821309890559797, "acc": 0.31555081370989124, "acc_stderr": 0.008003106736372202 }, "harness|drop|3": { "em": 0.003145973154362416, "em_stderr": 0.0005734993648436389, "f1": 0.04605390100671147, "f1_stderr": 0.0012821309890559797 }, "harness|gsm8k|5": { "acc": 0.0075815011372251705, "acc_stderr": 0.0023892815120772127 }, "harness|winogrande|5": { "acc": 0.6235201262825573, "acc_stderr": 0.01361693196066719 } }

配置详情

  • harness_arc_challenge_25

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|arc:challenge|25_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|arc:challenge|25_2023-09-03T08:57:35.818807.parquet
  • harness_drop_3

    • 分片:2023_10_16T04_13_51.306808
    • 路径:**/details_harness|drop|3_2023-10-16T04-13-51.306808.parquet
    • 分片:latest
    • 路径:**/details_harness|drop|3_2023-10-16T04-13-51.306808.parquet
  • harness_gsm8k_5

    • 分片:2023_10_16T04_13_51.306808
    • 路径:**/details_harness|gsm8k|5_2023-10-16T04-13-51.306808.parquet
    • 分片:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-16T04-13-51.306808.parquet
  • harness_hellaswag_10

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hellaswag|10_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hellaswag|10_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:多个文件路径,如**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:多个文件路径,如**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_anatomy_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_astronomy_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_college_biology_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_college_physics_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_computer_security_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_econometrics_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-03T08:57:35.818807.parquet
    • 分片:latest
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-09-03T08:57:35.818807.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分片:2023_09_03T08_57_35.818807
    • 路径:`**/details_harness|hendrycksTest-elementary_mathematics|
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