open-llm-leaderboard-old/details_Josephgflowers__Tinyllama-1.5B-Cinder-Test-4
收藏Hugging Face2024-04-06 更新2024-06-22 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard-old/details_Josephgflowers__Tinyllama-1.5B-Cinder-Test-4
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是在评估模型Josephgflowers/Tinyllama-1.5B-Cinder-Test-4时自动创建的,评估过程在Open LLM Leaderboard上进行。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,每次运行的结果作为特定配置中的一个分割,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,results配置存储了所有运行的聚合结果,用于在Open LLM Leaderboard上计算和显示聚合指标。
该数据集是在评估模型Josephgflowers/Tinyllama-1.5B-Cinder-Test-4时自动创建的,评估过程在Open LLM Leaderboard上进行。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,每次运行的结果作为特定配置中的一个分割,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,results配置存储了所有运行的聚合结果,用于在Open LLM Leaderboard上计算和显示聚合指标。
提供机构:
open-llm-leaderboard-old
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Josephgflowers/Tinyllama-1.5B-Cinder-Test-4在Open LLM Leaderboard上的自动创建的。数据集包含63个配置,每个配置对应一个评估任务。
数据集结构
数据集由1次运行创建,每个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。
额外配置
有一个额外的配置"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_Josephgflowers__Tinyllama-1.5B-Cinder-Test-4", "harness_winogrande_5", split="train")
最新结果
这些是最新结果来自运行2024-04-06T17:24:27.759280,包含各个任务的准确率(acc)、标准化准确率(acc_norm)、标准误差(acc_stderr、acc_norm_stderr)等指标。
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_04_06T17_24_27.759280
- path: **/details_harness|arc:challenge|25_2024-04-06T17-24-27.759280.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-04-06T17-24-27.759280.parquet
- split: 2024_04_06T17_24_27.759280
-
config_name: harness_gsm8k_5
- split: 2024_04_06T17_24_27.759280
- path: **/details_harness|gsm8k|5_2024-04-06T17-24-27.759280.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-04-06T17-24-27.759280.parquet
- split: 2024_04_06T17_24_27.759280
-
config_name: harness_hellaswag_10
- split: 2024_04_06T17_24_27.759280
- path: **/details_harness|hellaswag|10_2024-04-06T17-24-27.759280.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-04-06T17-24-27.759280.parquet
- split: 2024_04_06T17_24_27.759280
-
config_name: harness_hendrycksTest_5
- split: 2024_04_06T17_24_27.759280
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-04-06T17-24-27.759280.parquet
- **/details_harness|hendrycksTest-global_facts
- path:
- split: 2024_04_06T17_24_27.759280



