five

open-llm-leaderboard/details_RWKV__rwkv-raven-7b

收藏
Hugging Face2023-09-17 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_RWKV__rwkv-raven-7b
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RWKV/rwkv-raven-7b时自动生成的,主要用于在Open LLM Leaderboard上进行评估。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,并在Open LLM Leaderboard上显示聚合指标。
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 RWKV/rwkv-raven-7b 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集结构

  • 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,用于计算和显示聚合指标在 Open LLM Leaderboard 上。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_RWKV__rwkv-raven-7b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-17T20:05:27.789159 运行 的最新结果: python { "all": { "em": 0.10968959731543625, "em_stderr": 0.0032003177207929542, "f1": 0.15844798657718087, "f1_stderr": 0.0033868319824486323, "acc": 0.3160387943079502, "acc_stderr": 0.007545486731728054 }, "harness|drop|3": { "em": 0.10968959731543625, "em_stderr": 0.0032003177207929542, "f1": 0.15844798657718087, "f1_stderr": 0.0033868319824486323 }, "harness|gsm8k|5": { "acc": 0.003032600454890068, "acc_stderr": 0.0015145735612245395 }, "harness|winogrande|5": { "acc": 0.6290449881610103, "acc_stderr": 0.013576399902231568 } }

配置详情

  • harness_arc_challenge_25

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|arc:challenge|25_2023-09-03T10:19:01.730626.parquet
  • harness_drop_3

    • 分割: 2023_09_17T20_05_27.789159, latest
    • 路径: **/details_harness|drop|3_2023-09-17T20-05-27.789159.parquet
  • harness_gsm8k_5

    • 分割: 2023_09_17T20_05_27.789159, latest
    • 路径: **/details_harness|gsm8k|5_2023-09-17T20-05-27.789159.parquet
  • harness_hellaswag_10

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hellaswag|10_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: 多个路径,例如 **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-anatomy|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-astronomy|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-college_biology|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-college_physics|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-computer_security|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-econometrics|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-09-03T10:19:01.730626.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割: 2023_09_03T10_19_01.730626, latest
    • 路径: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-09-03T10:19:01.730626.parquet

以上是数据集的详细概述,包括数据集的简介、结构、加载示例、最新结果以及各个配置的详细信息。

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