open-llm-leaderboard-old/details_JunchengXie__Mistral-7B-Instruct-v0.1-gpt-4-80k-base_lora

该数据集是在评估模型[JunchengXie/Mistral-7B-Instruct-v0.1-gpt-4-80k-base_lora]时自动创建的，评估是在[Open LLM Leaderboard]上进行的。数据集包含63个配置，每个配置对应一个评估任务。数据集是从1次运行中创建的，每次运行都可以在特定配置中找到，分割名称使用运行的时间戳。此外，数据集还包含一个名为“results”的配置，用于存储所有运行的聚合结果，并用于计算和显示[Open LLM Leaderboard]上的聚合指标。

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 JunchengXie/Mistral-7B-Instruct-v0.1-gpt-4-80k-base_lora 在 Open LLM Leaderboard 上的运行过程中自动创建的。

数据集结构

• 配置数量：63个配置，每个配置对应一个评估任务。
• 数据来源：数据集来自1次运行，每个运行在每个配置中都有一个特定的分割，分割名称使用运行的时间戳。
• 特殊配置：额外配置 "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_JunchengXie__Mistral-7B-Instruct-v0.1-gpt-4-80k-base_lora", "harness_winogrande_5", split="train")

最新结果

以下是 2024-03-27T23:56:07.967023 运行的最新结果：

python { "all": { "acc": 0.5452277178528238, "acc_stderr": 0.03410493442579154, "acc_norm": 0.5519072949752738, "acc_norm_stderr": 0.03485960483749285, "mc1": 0.412484700122399, "mc1_stderr": 0.01723329939957122, "mc2": 0.5680735045864234, "mc2_stderr": 0.01559042793493668 }, "harness|arc:challenge|25": { "acc": 0.5162116040955631, "acc_stderr": 0.014603708567414945, "acc_norm": 0.5366894197952219, "acc_norm_stderr": 0.01457200052775699 }, "harness|hellaswag|10": { "acc": 0.5514837681736706, "acc_stderr": 0.004963259311700565, "acc_norm": 0.7358095996813384, "acc_norm_stderr": 0.004400000822742066 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.32, "acc_stderr": 0.046882617226215034, "acc_norm": 0.32, "acc_norm_stderr": 0.046882617226215034 }, ... }

配置详情

• harness_arc_challenge_25

  • 分割：2024_03_27T23_56_07.967023
  • 路径：**/details_harness|arc:challenge|25_2024-03-27T23-56-07.967023.parquet
  • 分割：latest
  • 路径：**/details_harness|arc:challenge|25_2024-03-27T23-56-07.967023.parquet

• harness_gsm8k_5

  • 分割：2024_03_27T23_56_07.967023
  • 路径：**/details_harness|gsm8k|5_2024-03-27T23-56-07.967023.parquet
  • 分割：latest
  • 路径：**/details_harness|gsm8k|5_2024-03-27T23-56-07.967023.parquet

• harness_hellaswag_10

  • 分割：2024_03_27T23_56_07.967023
  • 路径：**/details_harness|hellaswag|10_2024-03-27T23-56-07.967023.parquet
  • 分割：latest
  • 路径：**/details_harness|hellaswag|10_2024-03-27T23-56-07.967023.parquet

• harness_hendrycksTest_5

  • 分割：2024_03_27T23_56_07.967023
  • 路径：
    • **/details_harness|hendrycksTest-abstract_algebra|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-anatomy|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-astronomy|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-business_ethics|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-college_biology|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-college_chemistry|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-college_computer_science|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-college_mathematics|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-college_medicine|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-college_physics|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-computer_security|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-conceptual_physics|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-econometrics|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-electrical_engineering|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-27T23-56-07.967023.parquet
    • **/details_harness|hendrycksTest-formal_logic|5_2024-03-27T23-56-07.967023.parquet }