five

open-llm-leaderboard/details_KoboldAI__fairseq-dense-6.7B

收藏
Hugging Face2023-10-18 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_KoboldAI__fairseq-dense-6.7B
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在Open LLM Leaderboard上对模型KoboldAI/fairseq-dense-6.7B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。此外,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`函数加载数据集的示例。

This dataset was automatically created during the evaluation of the model KoboldAI/fairseq-dense-6.7B on the Open LLM Leaderboard. It consists of 64 configurations, each corresponding to a specific evaluation task. The dataset is generated from two runs, where each run is represented as a specific split for every configuration. The `train` split always points to the most recent results. Additionally, the configuration named `results` stores the aggregated results from all runs, which are used to calculate and display the aggregate metrics on the Open LLM Leaderboard. The README also provides an example of loading the dataset using the `load_dataset` function from the `datasets` library.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在评估模型 KoboldAI/fairseq-dense-6.7BOpen LLM Leaderboard 上的运行过程中自动创建的。

数据集结构

  • 配置数量:64个配置,每个配置对应一个评估任务。
  • 运行次数:数据集由2次运行创建。每次运行在每个配置中都有特定的分割,分割名称使用运行的时间戳。
  • 最新结果:"train" 分割始终指向最新的结果。
  • 汇总结果:一个额外的配置 "results" 存储所有运行的汇总结果,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_KoboldAI__fairseq-dense-6.7B", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-18T10:41:40.325981 运行的最新结果: python { "all": { "em": 0.08053691275167785, "em_stderr": 0.002786790295414266, "f1": 0.17048447986577175, "f1_stderr": 0.0032431717451241794, "acc": 0.32636148382004737, "acc_stderr": 0.006690454624875619 }, "harness|drop|3": { "em": 0.08053691275167785, "em_stderr": 0.002786790295414266, "f1": 0.17048447986577175, "f1_stderr": 0.0032431717451241794 }, "harness|gsm8k|5": { "acc": 0.0, "acc_stderr": 0.0 }, "harness|winogrande|5": { "acc": 0.6527229676400947, "acc_stderr": 0.013380909249751239 } }

配置详情

以下是部分配置的详细信息:

配置:harness_arc_challenge_25

  • 分割:2023_07_19T16_14_47.253287
    • 路径
      • **/details_harness|arc:challenge|25_2023-07-19T16:14:47.253287.parquet
  • 分割:latest
    • 路径
      • **/details_harness|arc:challenge|25_2023-07-19T16:14:47.253287.parquet

配置:harness_drop_3

  • 分割:2023_10_18T10_41_40.325981
    • 路径
      • **/details_harness|drop|3_2023-10-18T10-41-40.325981.parquet
  • 分割:latest
    • 路径
      • **/details_harness|drop|3_2023-10-18T10-41-40.325981.parquet

配置:harness_gsm8k_5

  • 分割:2023_10_18T10_41_40.325981
    • 路径
      • **/details_harness|gsm8k|5_2023-10-18T10-41-40.325981.parquet
  • 分割:latest
    • 路径
      • **/details_harness|gsm8k|5_2023-10-18T10-41-40.325981.parquet

配置:harness_hellaswag_10

  • 分割:2023_07_19T16_14_47.253287
    • 路径
      • **/details_harness|hellaswag|10_2023-07-19T16:14:47.253287.parquet
  • 分割:latest
    • 路径
      • **/details_harness|hellaswag|10_2023-07-19T16:14:47.253287.parquet

配置:harness_hendrycksTest_5

  • 分割:2023_07_19T16_14_47.253287
    • 路径
      • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-anatomy|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-astronomy|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-college_biology|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-college_physics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-computer_security|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-econometrics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-global_facts|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-human_aging|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-international_law|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-management|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-marketing|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-nutrition|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-philosophy|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-prehistory|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-19T16:14:47.253287.parquet
      • **/details_harness|hendrycksTest-professional_law|5_2023-07-19T16:14:47.2
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