five

open-llm-leaderboard-old/details_bofenghuang__vigogne-13b-instruct

收藏
Hugging Face2023-09-23 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_bofenghuang__vigogne-13b-instruct
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bofenghuang/vigogne-13b-instruct在Open LLM Leaderboard上的评估运行期间自动创建的。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。train分割始终指向最新的结果。一个额外的配置results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还提供了如何使用Python中的datasets库加载运行细节的示例。

该数据集是在模型bofenghuang/vigogne-13b-instruct在Open LLM Leaderboard上的评估运行期间自动创建的。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。train分割始终指向最新的结果。一个额外的配置results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还提供了如何使用Python中的datasets库加载运行细节的示例。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

该数据集是在对模型 bofenghuang/vigogne-13b-instruct 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。

数据集结构

数据集包含以下配置:

  • harness_arc_challenge_25

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|arc:challenge|25_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-07-24T14:44:50.658520.parquet
  • harness_drop_3

    • 分割:2023_09_23T12_07_22.235368
    • 路径:**/details_harness|drop|3_2023-09-23T12-07-22.235368.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-09-23T12-07-22.235368.parquet
  • harness_gsm8k_5

    • 分割:2023_09_23T12_07_22.235368
    • 路径:**/details_harness|gsm8k|5_2023-09-23T12-07-22.235368.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-09-23T12-07-22.235368.parquet
  • harness_hellaswag_10

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hellaswag|10_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:多个文件路径
    • 分割:latest
    • 路径:多个文件路径
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-24T14:44:50.658520.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割:2023_07_24T14_44_50.658520
    • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-24T14:44:50.658520.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-24T14:44:50.658520.parquet

最新结果

以下是2023年9月23日12:07:22.235368运行的最新结果:

python { "all": { "em": 0.17365771812080538, "em_stderr": 0.003879418958892452, "f1": 0.24381711409395945, "f1_stderr": 0.003948980378514697, "acc": 0.44469214138811486, "acc_stderr": 0.010351218038230168 }, "harness|drop|3": { "em": 0.17365771812080538, "em_stderr": 0.003879418958892452, "f1": 0.24381711409395945, "f1_stderr": 0.003948980378514697 }, "harness|gsm8k|5": { "acc": 0.11827141774071266, "acc_stderr": 0.008895075852434951 }, "harness|winogrande|5": { "acc": 0.771112865035517, "acc_stderr": 0.011807360224025386 } }

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