five

open-llm-leaderboard-old/details_euclaise__falcon_1b_stage3_2

收藏
Hugging Face2023-10-24 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_euclaise__falcon_1b_stage3_2
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 euclaise/falcon_1b_stage3_2 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,每个配置对应一个特定的评估任务。数据集是从 2 次运行中生成的,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。一个名为 results 的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个示例,展示了如何使用 `datasets` 库中的 `load_dataset` 函数加载数据集。

该数据集是在模型 euclaise/falcon_1b_stage3_2 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,每个配置对应一个特定的评估任务。数据集是从 2 次运行中生成的,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。一个名为 results 的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个示例,展示了如何使用 `datasets` 库中的 `load_dataset` 函数加载数据集。
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 euclaise/falcon_1b_stage3_2 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

数据集由以下配置组成:

  • harness_arc_challenge_25

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|arc:challenge|25_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-10-10T19-39-31.631601.parquet
  • harness_drop_3

    • 分割:2023_10_24T01_21_10.323077
    • 路径:**/details_harness|drop|3_2023-10-24T01-21-10.323077.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-10-24T01-21-10.323077.parquet
  • harness_gsm8k_5

    • 分割:2023_10_24T01_21_10.323077
    • 路径:**/details_harness|gsm8k|5_2023-10-24T01-21-10.323077.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-10-24T01-21-10.323077.parquet
  • harness_hellaswag_10

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hellaswag|10_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-10-10T19-39-31.631601.parquet
  • harness_hendrycksTest_elementary_mathematics_5

    • 分割:2023_10_10T19_39_31.631601
    • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-10T19-39-31.631601.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-10-10T19-39-31.631601.parquet

最新结果

以下是2023年10月24日01:21:10.323077运行的最新结果:

python { "all": { "em": 0.15541107382550334, "em_stderr": 0.0037102512781803346, "f1": 0.2052170721476513, "f1_stderr": 0.0038511842764741527, "acc": 0.3022888713496448, "acc_stderr": 0.006870839193772672 }, "harness|drop|3": { "em": 0.15541107382550334, "em_stderr": 0.0037102512781803346, "f1": 0.2052170721476513, "f1_stderr": 0.0038511842764741527 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|winogrande|5": { "acc": 0.6045777426992897, "acc_stderr": 0.013741678387545343 } }

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