open-llm-leaderboard-old/details_TheBloke__neural-chat-7B-v3-2-GPTQ

该数据集是在评估模型TheBloke/neural-chat-7B-v3-2-GPTQ时自动创建的，用于在Open LLM Leaderboard上进行评估。数据集包含63个配置，每个配置对应一个评估任务。数据集由1次运行生成，每次运行的结果可以在每个配置的特定分割中找到，分割名称使用运行的时间戳命名。train分割始终指向最新的结果。此外，一个名为results的配置存储了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

该数据集是在对模型 TheBloke/neural-chat-7B-v3-2-GPTQ 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 63 个配置，每个配置对应一个评估任务。
• 数据集从 1 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 Open LLM Leaderboard 上的聚合指标。

最新结果

• 这些是最新的结果，来自 2023-12-11T00:12:21.907526 的运行。
• 结果包括多个任务的准确率（acc）、标准误差（acc_stderr）、归一化准确率（acc_norm）和归一化标准误差（acc_norm_stderr）等指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_TheBloke__neural-chat-7B-v3-2-GPTQ", "harness_winogrande_5", split="train")

配置详情

• harness_arc_challenge_25

  • 分割：2023_12_11T00_12_21.907526, latest
  • 路径：**/details_harness|arc:challenge|25_2023-12-11T00-12-21.907526.parquet

• harness_gsm8k_5

  • 分割：2023_12_11T00_12_21.907526, latest
  • 路径：**/details_harness|gsm8k|5_2023-12-11T00-12-21.907526.parquet

• harness_hellaswag_10

  • 分割：2023_12_11T00_12_21.907526, latest
  • 路径：**/details_harness|hellaswag|10_2023-12-11T00-12-21.907526.parquet

• harness_hendrycksTest_5

  • 分割：2023_12_11T00_12_21.907526
  • 路径：多个文件路径，包括 abstract_algebra, anatomy, astronomy, business_ethics, clinical_knowledge, college_biology, college_chemistry, college_computer_science, college_mathematics, college_medicine, college_physics, computer_security, conceptual_physics, econometrics, electrical_engineering, elementary_mathematics, formal_logic, global_facts, high_school_biology, high_school_chemistry, high_school_computer_science, high_school_european_history, high_school_geography, high_school_government_and_politics, high_school_macroeconomics, high_school_mathematics, high_school_microeconomics, high_school_physics, high_school_psychology, high_school_statistics, high_school_us_history, high_school_world_history, human_aging, human_sexuality, international_law, jurisprudence, logical_fallacies, machine_learning, management, marketing, medical_genetics, miscellaneous, moral_disputes, moral_scenarios, nutrition, philosophy, prehistory, professional_accounting, professional_law, professional_medicine, professional_psychology, public_relations, security_studies, sociology, us_foreign_policy, virology, world_religions 等任务的详细文件。