open-llm-leaderboard-old/details_chlee10__T3Q-Platypus-Mistral7B
收藏数据集概述
该数据集是在评估模型chlee10/T3Q-Platypus-Mistral7B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
- 配置数量:63个配置
- 数据来源:从1次运行中创建,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。
- 最新结果:"train"分割始终指向最新结果。
- 结果汇总:一个额外的配置"results"存储所有运行的汇总结果,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_chlee10__T3Q-Platypus-Mistral7B", "harness_winogrande_5", split="train")
最新结果
以下是最新结果来自2024-03-12T18:02:29.426702的摘要:
python { "all": { "acc": 0.610489556078658, "acc_stderr": 0.03291521797225751, "acc_norm": 0.6115205749524352, "acc_norm_stderr": 0.03359123137175418, "mc1": 0.3488372093023256, "mc1_stderr": 0.01668441985998689, "mc2": 0.5185430742867603, "mc2_stderr": 0.015260226649004952 }, "harness|arc:challenge|25": { "acc": 0.591296928327645, "acc_stderr": 0.014365750345427001, "acc_norm": 0.6313993174061433, "acc_norm_stderr": 0.0140978106780422 }, "harness|hellaswag|10": { "acc": 0.6437960565624378, "acc_stderr": 0.0047789780313896415, "acc_norm": 0.8440549691296555, "acc_norm_stderr": 0.003620617550747393 }, ... }
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分割:2024_03_12T18_02_29.426702
- 路径:
**/details_harness|arc:challenge|25_2024-03-12T18-02-29.426702.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2024-03-12T18-02-29.426702.parquet
- 路径:
- 分割:2024_03_12T18_02_29.426702
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分割:2024_03_12T18_02_29.426702
- 路径:
**/details_harness|gsm8k|5_2024-03-12T18-02-29.426702.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2024-03-12T18-02-29.426702.parquet
- 路径:
- 分割:2024_03_12T18_02_29.426702
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分割:2024_03_12T18_02_29.426702
- 路径:
**/details_harness|hellaswag|10_2024-03-12T18-02-29.426702.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2024-03-12T18-02-29.426702.parquet
- 路径:
- 分割:2024_03_12T18_02_29.426702
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分割:2024_03_12T18_02_29.426702
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-anatomy|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-astronomy|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_biology|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_physics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-computer_security|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-econometrics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-global_facts|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-03-12T18-02-29.426702.parquet- ...
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-anatomy|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-astronomy|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_biology|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-college_physics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-computer_security|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-econometrics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-global_facts|5_2024-03-12T18-02-29.426702.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-03-12T18-02-29.426702.parquet- ...
- 路径:
- 分割:2024_03_12T18_02_29.426702
- 数据文件:



