open-llm-leaderboard-old/details_Sao10K__Winterreise-m7
收藏数据集概述
数据集简介
该数据集是在评估模型 Sao10K/Winterreise-m7 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
数据集从 1 次运行中创建,每个运行可以在每个配置中找到一个特定的分割,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_Sao10K__Winterreise-m7", "harness_winogrande_5", split="train")
最新结果
以下是 2024-01-06T17:27:54.681124 运行的最新结果:
python { "all": { "acc": 0.6356608689538731, "acc_stderr": 0.03240626820159796, "acc_norm": 0.6414323310244244, "acc_norm_stderr": 0.03306251040203467, "mc1": 0.30354957160342716, "mc1_stderr": 0.016095884155386847, "mc2": 0.4555071503037089, "mc2_stderr": 0.014269000225909255 }, "harness|arc:challenge|25": { "acc": 0.5784982935153583, "acc_stderr": 0.014430197069326023, "acc_norm": 0.6126279863481229, "acc_norm_stderr": 0.01423587248790987 }, "harness|hellaswag|10": { "acc": 0.6364270065723959, "acc_stderr": 0.004800446397653348, "acc_norm": 0.8383788090021908, "acc_norm_stderr": 0.0036735065123709503 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.36, "acc_stderr": 0.04824181513244218, "acc_norm": 0.36, "acc_norm_stderr": 0.04824181513244218 }, # 其他任务的结果省略... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_01_06T17_27_54.681124
- path: **/details_harness|arc:challenge|25_2024-01-06T17-27-54.681124.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-01-06T17-27-54.681124.parquet
- split: 2024_01_06T17_27_54.681124
-
config_name: harness_gsm8k_5
- split: 2024_01_06T17_27_54.681124
- path: **/details_harness|gsm8k|5_2024-01-06T17-27-54.681124.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-01-06T17-27-54.681124.parquet
- split: 2024_01_06T17_27_54.681124
-
config_name: harness_hellaswag_10
- split: 2024_01_06T17_27_54.681124
- path: **/details_harness|hellaswag|10_2024-01-06T17-27-54.681124.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-01-06T17-27-54.681124.parquet
- split: 2024_01_06T17_27_54.681124
-
config_name: harness_hendrycksTest_5
- split: 2024_01_06T17_27_54.681124
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-global_facts|5_2024-01-06T17-27-54.681124.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2024-01-06T17-27-54.681124.parquet
其他路径省略...
- path:
- split: 2024_01_06T17_27_54.681124
}



