five

open-llm-leaderboard/details_TheBloke__Nous-Hermes-13B-SuperHOT-8K-fp16

收藏
Hugging Face2023-10-22 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_TheBloke__Nous-Hermes-13B-SuperHOT-8K-fp16
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TheBloke/Nous-Hermes-13B-SuperHOT-8K-fp16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。数据集由2次运行生成,每次运行在每个配置中表示为特定的分割,分割名称使用运行的时间戳。train分割始终指向最新结果。此外,名为results的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还提供了如何使用`datasets`库中的`load_dataset`函数加载运行详情的示例。

This dataset was automatically created during the evaluation run of the model TheBloke/Nous-Hermes-13B-SuperHOT-8K-fp16 on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to a specific evaluation task. The dataset is generated from 2 runs, where each run is represented as a specific data split under each configuration, with the split name using the timestamp of the run. 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 aggregated metrics on the Open LLM Leaderboard. The README also provides examples of how to use the `load_dataset` function from the `datasets` library to load the run details.
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 TheBloke/Nous-Hermes-13B-SuperHOT-8K-fp16Open LLM Leaderboard 上的自动创建的。

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。
  • 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_TheBloke__Nous-Hermes-13B-SuperHOT-8K-fp16", "harness_winogrande_5", split="train")

最新结果

以下是 最新结果来自 run 2023-10-22T21:24:49.496203 的摘要: python { "all": { "em": 0.24779781879194632, "em_stderr": 0.004421358038007316, "f1": 0.3203208892617463, "f1_stderr": 0.004418252169927022, "acc": 0.3825450746272229, "acc_stderr": 0.007568348592873263 }, "harness|drop|3": { "em": 0.24779781879194632, "em_stderr": 0.004421358038007316, "f1": 0.3203208892617463, "f1_stderr": 0.004418252169927022 }, "harness|gsm8k|5": { "acc": 0.012130401819560273, "acc_stderr": 0.003015294242890953 }, "harness|winogrande|5": { "acc": 0.7529597474348856, "acc_stderr": 0.012121402942855573 } }

配置详情

以下是数据集的配置详情:

  • harness_arc_challenge_25

    • 分割: 2023_08_01T13_07_54.585648
      • 路径: **/details_harness|arc:challenge|25_2023-08-01T13:07:54.585648.parquet
    • 分割: latest
      • 路径: **/details_harness|arc:challenge|25_2023-08-01T13:07:54.585648.parquet
  • harness_drop_3

    • 分割: 2023_10_22T21_24_49.496203
      • 路径: **/details_harness|drop|3_2023-10-22T21-24-49.496203.parquet
    • 分割: latest
      • 路径: **/details_harness|drop|3_2023-10-22T21-24-49.496203.parquet
  • harness_gsm8k_5

    • 分割: 2023_10_22T21_24_49.496203
      • 路径: **/details_harness|gsm8k|5_2023-10-22T21-24-49.496203.parquet
    • 分割: latest
      • 路径: **/details_harness|gsm8k|5_2023-10-22T21-24-49.496203.parquet
  • harness_hellaswag_10

    • 分割: 2023_08_01T13_07_54.585648
      • 路径: **/details_harness|hellaswag|10_2023-08-01T13:07:54.585648.parquet
    • 分割: latest
      • 路径: **/details_harness|hellaswag|10_2023-08-01T13:07:54.585648.parquet
  • harness_hendrycksTest_5

    • 分割: 2023_08_01T13_07_54.585648
      • 路径:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-01T13:07:54.585648.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-01T13:07:54.585648.parquet
        • `**/details_harness|hendrycksTest-professional_
二维码
社区交流群
二维码
科研交流群
商业服务