open-llm-leaderboard-old/details_KnutJaegersberg__Walter-Mistral-7B
收藏数据集概述
数据集简介
该数据集是在评估模型 KnutJaegersberg/Walter-Mistral-7B 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
一个额外的配置 "results" 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。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_KnutJaegersberg__Walter-Mistral-7B", "harness_winogrande_5", split="train")
最新结果
以下是 2023-12-18T13:56:54.383446 运行的最新结果:
python { "all": { "acc": 0.5800043815414543, "acc_stderr": 0.03295379522396862, "acc_norm": 0.590722838232918, "acc_norm_stderr": 0.03383616633951902, "mc1": 0.2668298653610771, "mc1_stderr": 0.015483691939237265, "mc2": 0.399340955094558, "mc2_stderr": 0.013888761310440584 }, "harness|arc:challenge|25": { "acc": 0.5469283276450512, "acc_stderr": 0.014546892052005626, "acc_norm": 0.5887372013651877, "acc_norm_stderr": 0.01437944106852208 }, "harness|hellaswag|10": { "acc": 0.6326428998207528, "acc_stderr": 0.0048109966523247295, "acc_norm": 0.8342959569806812, "acc_norm_stderr": 0.0037105487209054206 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.32, "acc_stderr": 0.04688261722621503, "acc_norm": 0.32, "acc_norm_stderr": 0.04688261722621503 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_12_18T13_56_54.383446
- path: **/details_harness|arc:challenge|25_2023-12-18T13-56-54.383446.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-12-18T13-56-54.383446.parquet
- split: 2023_12_18T13_56_54.383446
-
config_name: harness_gsm8k_5
- split: 2023_12_18T13_56_54.383446
- path: **/details_harness|gsm8k|5_2023-12-18T13-56-54.383446.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-12-18T13-56-54.383446.parquet
- split: 2023_12_18T13_56_54.383446
-
config_name: harness_hellaswag_10
- split: 2023_12_18T13_56_54.383446
- path: **/details_harness|hellaswag|10_2023-12-18T13-56-54.383446.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-12-18T13-56-54.383446.parquet
- split: 2023_12_18T13_56_54.383446
-
config_name: harness_hendrycksTest_5
- split: 2023_12_18T13_56_54.383446
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-12-18T13-56-54.383446.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-12-18T13-56-54.383446.parquet
其他任务的路径...
- path:
- split: 2023_12_18T13_56_54.383446
}



