five

open-llm-leaderboard-old/details_danielhanchen__open_llama_3b_600bt_preview

收藏
Hugging Face2023-09-22 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_danielhanchen__open_llama_3b_600bt_preview
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danielhanchen/open_llama_3b_600bt_preview在Open LLM Leaderboard上进行评估时自动生成的。数据集包含64个配置,每个配置对应一个评估任务。它包含2次运行的数据,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新结果。此外,results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Python代码片段用于加载数据集,并显示了特定运行的最新结果。

该数据集是在模型danielhanchen/open_llama_3b_600bt_preview在Open LLM Leaderboard上进行评估时自动生成的。数据集包含64个配置,每个配置对应一个评估任务。它包含2次运行的数据,每次运行在每个配置中表示为特定的分割,分割名称由运行的时间戳命名。train分割始终指向最新结果。此外,results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Python代码片段用于加载数据集,并显示了特定运行的最新结果。
提供机构:
open-llm-leaderboard-old
原始信息汇总

数据集概述

数据集是在评估模型 danielhanchen/open_llama_3b_600bt_previewOpen LLM Leaderboard 上的运行过程中自动创建的。

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_danielhanchen__open_llama_3b_600bt_preview", "harness_winogrande_5", split="train")

最新结果

以下是 2023-09-22T13:47:34.979572 运行的最新结果: python { "all": { "em": 0.0009437919463087249, "em_stderr": 0.0003144653119413175, "f1": 0.04996329697986588, "f1_stderr": 0.0012567293128089149, "acc": 0.32150142444857593, "acc_stderr": 0.007826931083969837 }, "harness|drop|3": { "em": 0.0009437919463087249, "em_stderr": 0.0003144653119413175, "f1": 0.04996329697986588, "f1_stderr": 0.0012567293128089149 }, "harness|gsm8k|5": { "acc": 0.006065200909780136, "acc_stderr": 0.002138670301460455 }, "harness|winogrande|5": { "acc": 0.6369376479873717, "acc_stderr": 0.01351519186647922 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|arc:challenge|25_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|arc:challenge|25_2023-07-19T15:00:20.394414.parquet
  • harness_drop_3

    • 分割:2023_09_22T13_47_34.979572
    • 路径:**/details_harness|drop|3_2023-09-22T13-47-34.979572.parquet
    • 分割:latest
    • 路径:**/details_harness|drop|3_2023-09-22T13-47-34.979572.parquet
  • harness_gsm8k_5

    • 分割:2023_09_22T13_47_34.979572
    • 路径:**/details_harness|gsm8k|5_2023-09-22T13-47-34.979572.parquet
    • 分割:latest
    • 路径:**/details_harness|gsm8k|5_2023-09-22T13-47-34.979572.parquet
  • harness_hellaswag_10

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hellaswag|10_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hellaswag|10_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:00:20.394414.parquet 等 40 个文件
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:00:20.394414.parquet 等 40 个文件
  • harness_hendrycksTest_abstract_algebra_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_anatomy_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_astronomy_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_business_ethics_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_clinical_knowledge_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_college_biology_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_college_chemistry_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_college_computer_science_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_college_mathematics_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_college_medicine_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_college_physics_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_computer_security_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_conceptual_physics_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_econometrics_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-07-19T15:00:20.394414.parquet
    • 分割:latest
    • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-07-19T15:00:20.394414.parquet
  • harness_hendrycksTest_electrical_engineering_5

    • 分割:2023_07_19T15_00_20.394414
    • 路径:`**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T15:00:20.3
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