five

open-llm-leaderboard/details_jondurbin__airoboros-l2-70b-gpt4-m2.0

收藏
Hugging Face2023-10-23 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_jondurbin__airoboros-l2-70b-gpt4-m2.0
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 jondurbin/airoboros-l2-70b-gpt4-m2.0 在 Open LLM Leaderboard 上的评估运行期间自动生成的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。数据集是从 3 次运行中创建的,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 分割始终指向最新的结果。此外,还有一个 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 还提供了如何使用 Python 代码加载运行中的数据集详细信息的示例,并包含了特定运行的最新结果。

This dataset was automatically generated during the evaluation run of the model jondurbin/airoboros-l2-70b-gpt4-m2.0 on the Open LLM Leaderboard. It consists of 64 configurations, each corresponding to one evaluated task. The dataset is created from 3 runs, where each run is represented as a specific split under each configuration, with the split name using the timestamp of the run. The 'train' split always points to the most recent results. In addition, there is a 'results' configuration that 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 examples of how to use Python code to load detailed information about the runs, and includes the most recent results from specific runs.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集名称

Evaluation run of jondurbin/airoboros-l2-70b-gpt4-m2.0

数据集摘要

该数据集是在模型jondurbin/airoboros-l2-70b-gpt4-m2.0Open LLM Leaderboard上的评估运行期间自动创建的。

数据集组成

  • 数据集由64个配置组成,每个配置对应一个评估任务。
  • 数据集从3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_jondurbin__airoboros-l2-70b-gpt4-m2.0", "harness_winogrande_5", split="train")

最新结果

以下是2023-10-23T01:01:59.537351运行的最新结果: python { "all": { "em": 0.19955956375838926, "em_stderr": 0.004092987650196818, "f1": 0.2619746224832211, "f1_stderr": 0.004049847572493045, "acc": 0.5449064818543622, "acc_stderr": 0.011200400992385444 }, "harness|drop|3": { "em": 0.19955956375838926, "em_stderr": 0.004092987650196818, "f1": 0.2619746224832211, "f1_stderr": 0.004049847572493045 }, "harness|gsm8k|5": { "acc": 0.2539802880970432, "acc_stderr": 0.011989952209548084 }, "harness|winogrande|5": { "acc": 0.8358326756116812, "acc_stderr": 0.010410849775222804 } }

配置详情

  • harness_arc_challenge_25

    • 分割:2023_08_09T22_06_19.540113
      • 路径:**/details_harness|arc:challenge|25_2023-08-09T22:06:19.540113.parquet
    • 分割:2023_08_10T00_00_29.305175
      • 路径:**/details_harness|arc:challenge|25_2023-08-10T00:00:29.305175.parquet
    • 分割:latest
      • 路径:**/details_harness|arc:challenge|25_2023-08-10T00:00:29.305175.parquet
  • harness_drop_3

    • 分割:2023_10_23T01_01_59.537351
      • 路径:**/details_harness|drop|3_2023-10-23T01-01-59.537351.parquet
    • 分割:latest
      • 路径:**/details_harness|drop|3_2023-10-23T01-01-59.537351.parquet
  • harness_gsm8k_5

    • 分割:2023_10_23T01_01_59.537351
      • 路径:**/details_harness|gsm8k|5_2023-10-23T01-01-59.537351.parquet
    • 分割:latest
      • 路径:**/details_harness|gsm8k|5_2023-10-23T01-01-59.537351.parquet
  • harness_hellaswag_10

    • 分割:2023_08_09T22_06_19.540113
      • 路径:**/details_harness|hellaswag|10_2023-08-09T22:06:19.540113.parquet
    • 分割:2023_08_10T00_00_29.305175
      • 路径:**/details_harness|hellaswag|10_2023-08-10T00:00:29.305175.parquet
    • 分割:latest
      • 路径:**/details_harness|hellaswag|10_2023-08-10T00:00:29.305175.parquet
  • harness_hendrycksTest_5

    • 分割:2023_08_09T22_06_19.540113
      • 路径:**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-anatomy|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-astronomy|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-college_biology|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-college_physics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-computer_security|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-econometrics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-formal_logic|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-global_facts|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_biology|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_european_history|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_geography|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_physics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_psychology|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_statistics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_us_history|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-high_school_world_history|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-human_aging|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-human_sexuality|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-international_law|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-jurisprudence|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-logical_fallacies|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-machine_learning|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-management|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-marketing|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-medical_genetics|5_2023-08-09T22:06:19.540113.parquet
      • 路径:**/details_harness|hendrycksTest-miscellaneous|5_2023-08-09T22:06:19.540113.parquet
      • 路径:`**/details_harness|hendrycksTest-moral_disputes|5_2023-08
二维码
社区交流群
二维码
科研交流群
商业服务