five

open-llm-leaderboard/details_Norquinal__llama-2-7b-claude-chat

收藏
Hugging Face2023-09-17 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_Norquinal__llama-2-7b-claude-chat
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 Norquinal/llama-2-7b-claude-chat 在 Open LLM Leaderboard 上评估期间自动创建的。数据集包含 64 个配置,每个配置对应一个被评估的任务。数据集由 2 次运行的数据组成,每次运行的数据存储为以时间戳命名的特定分割。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 上的指标。README 还提供了如何使用 Python 中的 datasets 库加载数据集的示例,并包含了特定运行的最新结果。

This dataset was automatically created during the evaluation of the model Norquinal/llama-2-7b-claude-chat on the Open LLM Leaderboard. The dataset consists of 64 configurations, each corresponding to one evaluated task. The dataset comprises data from 2 evaluation runs, with data from each run stored as a specific split named after a timestamp. The `train` split always points to the most recent results. In addition, the configuration named `results` stores the aggregated results from all runs, which are used to calculate and display the metrics on the Open LLM Leaderboard. The README also provides examples of how to load the dataset using the `datasets` library in Python, and includes the most recent results for specific runs.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集摘要

该数据集是在对模型Norquinal/llama-2-7b-claude-chat进行评估运行期间自动创建的,用于Open LLM Leaderboard

数据集组成

  • 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_Norquinal__llama-2-7b-claude-chat", "harness_winogrande_5", split="train")

最新结果

以下是2023-09-17T21:50:18.954049运行的最新结果: python { "all": { "em": 0.0016778523489932886, "em_stderr": 0.00041913301788269335, "f1": 0.0588590604026846, "f1_stderr": 0.0013399207626085948, "acc": 0.4131723645607008, "acc_stderr": 0.009771744871869251 }, "harness|drop|3": { "em": 0.0016778523489932886, "em_stderr": 0.00041913301788269335, "f1": 0.0588590604026846, "f1_stderr": 0.0013399207626085948 }, "harness|gsm8k|5": { "acc": 0.07733131159969674, "acc_stderr": 0.00735771352322235 }, "harness|winogrande|5": { "acc": 0.7490134175217048, "acc_stderr": 0.012185776220516153 } }

配置详情

  • config_name: harness_arc_challenge_25

    • split: 2023_08_30T02_05_21.068223
      • path: **/details_harness|arc:challenge|25_2023-08-30T02:05:21.068223.parquet
    • split: latest
      • path: **/details_harness|arc:challenge|25_2023-08-30T02:05:21.068223.parquet
  • config_name: harness_drop_3

    • split: 2023_09_17T21_50_18.954049
      • path: **/details_harness|drop|3_2023-09-17T21-50-18.954049.parquet
    • split: latest
      • path: **/details_harness|drop|3_2023-09-17T21-50-18.954049.parquet
  • config_name: harness_gsm8k_5

    • split: 2023_09_17T21_50_18.954049
      • path: **/details_harness|gsm8k|5_2023-09-17T21-50-18.954049.parquet
    • split: latest
      • path: **/details_harness|gsm8k|5_2023-09-17T21-50-18.954049.parquet
  • config_name: harness_hellaswag_10

    • split: 2023_08_30T02_05_21.068223
      • path: **/details_harness|hellaswag|10_2023-08-30T02:05:21.068223.parquet
    • split: latest
      • path: **/details_harness|hellaswag|10_2023-08-30T02:05:21.068223.parquet
  • config_name: harness_hendrycksTest_5

    • split: 2023_08_30T02_05_21.068223
      • path:
        • **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-anatomy|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-astronomy|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-business_ethics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-college_biology|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-college_chemistry|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-college_computer_science|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-college_mathematics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-college_medicine|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-college_physics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-computer_security|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-econometrics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-formal_logic|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-global_facts|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_biology|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_geography|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_physics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-human_aging|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-human_sexuality|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-international_law|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-jurisprudence|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-machine_learning|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-management|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-marketing|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-medical_genetics|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-miscellaneous|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-moral_disputes|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-nutrition|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-philosophy|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-prehistory|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-professional_accounting|5_2023-08-30T02:05:21.068223.parquet
        • **/details_harness|hendrycksTest-professional_law|5_2023-08-30T02:05:21.068223.parquet
二维码
社区交流群
二维码
科研交流群
商业服务