open-llm-leaderboard-old/details_cloudyu__Phoenix_DPO_60B
收藏数据集概述
该数据集是在评估模型cloudyu/Phoenix_DPO_60B在Open LLM Leaderboard上的运行过程中自动创建的。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一个特定的分割,分割名称使用运行的时间戳。
- 最新结果:"train"分割总是指向最新的结果。
- 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_cloudyu__Phoenix_DPO_60B", "harness_winogrande_5", split="train")
最新结果
以下是最新结果从运行2024-01-29T07:18:26.493020的摘要:
python { "all": { "acc": 0.773060355126592, "acc_stderr": 0.02788677122058737, "acc_norm": 0.7768613327892654, "acc_norm_stderr": 0.028421288853019304, "mc1": 0.47613219094247244, "mc1_stderr": 0.017483547156961574, "mc2": 0.6383841795252835, "mc2_stderr": 0.014733348951101679 }, "harness|arc:challenge|25": { "acc": 0.6877133105802048, "acc_stderr": 0.013542598541688067, "acc_norm": 0.71160409556314, "acc_norm_stderr": 0.013238394422428173 }, "harness|hellaswag|10": { "acc": 0.6528579964150567, "acc_stderr": 0.004750884401095161, "acc_norm": 0.8546106353316073, "acc_norm_stderr": 0.003517725787017736 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.52, "acc_stderr": 0.050211673156867795, "acc_norm": 0.52, "acc_norm_stderr": 0.050211673156867795 }, ... }
配置详情
-
配置名称:harness_arc_challenge_25
- 数据文件:
- 分割:2024_01_29T07_18_26.493020
- 路径:
**/details_harness|arc:challenge|25_2024-01-29T07-18-26.493020.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2024-01-29T07-18-26.493020.parquet
- 路径:
- 分割:2024_01_29T07_18_26.493020
- 数据文件:
-
配置名称:harness_gsm8k_5
- 数据文件:
- 分割:2024_01_29T07_18_26.493020
- 路径:
**/details_harness|gsm8k|5_2024-01-29T07-18-26.493020.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|gsm8k|5_2024-01-29T07-18-26.493020.parquet
- 路径:
- 分割:2024_01_29T07_18_26.493020
- 数据文件:
-
配置名称:harness_hellaswag_10
- 数据文件:
- 分割:2024_01_29T07_18_26.493020
- 路径:
**/details_harness|hellaswag|10_2024-01-29T07-18-26.493020.parquet
- 路径:
- 分割:latest
- 路径:
**/details_harness|hellaswag|10_2024-01-29T07-18-26.493020.parquet
- 路径:
- 分割:2024_01_29T07_18_26.493020
- 数据文件:
-
配置名称:harness_hendrycksTest_5
- 数据文件:
- 分割:2024_01_29T07_18_26.493020
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-29T07-18-26.493020.parquet- ...
- 路径:
- 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-anatomy|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-astronomy|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_biology|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-college_physics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-computer_security|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-econometrics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-global_facts|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-01-29T07-18-26.493020.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-01-29T07-18-26.493020.parquet- ... }
- 路径:
- 分割:2024_01_29T07_18_26.493020
- 数据文件:



