open-llm-leaderboard/details_Salesforce__codegen-16B-nl
收藏数据集概述
数据集简介
该数据集是在对模型 Salesforce/codegen-16B-nl 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
数据集组成
- 数据集包含 122 个配置,每个配置对应一个评估任务。
- 数据集从 3 次运行中创建,每次运行可以在每个配置中找到特定的拆分,拆分名称使用运行的时间戳。
- "train" 拆分始终指向最新的结果。
- 额外的 "results" 配置存储所有运行的聚合结果,用于计算和显示聚合指标在 Open LLM Leaderboard 上。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_Salesforce__codegen-16B-nl", "harness_winogrande_5", split="train")
最新结果
以下是 2023-09-16T21:34:04.587307 运行的最新结果: python { "all": { "em": 0.0012583892617449664, "em_stderr": 0.00036305608931190987, "f1": 0.050115352348993385, "f1_stderr": 0.0012004040103300048, "acc": 0.35304663251500595, "acc_stderr": 0.00877106572247344 }, "harness|drop|3": { "em": 0.0012583892617449664, "em_stderr": 0.00036305608931190987, "f1": 0.050115352348993385, "f1_stderr": 0.0012004040103300048 }, "harness|gsm8k|5": { "acc": 0.026535253980288095, "acc_stderr": 0.004427045987265165 }, "harness|winogrande|5": { "acc": 0.6795580110497238, "acc_stderr": 0.013115085457681714 } }
配置详情
-
harness_arc_challenge_25
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|arc:challenge|25_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|arc:challenge|25_2023-07-19T20:43:28.351317.parquet
-
harness_drop_3
- 拆分:2023_09_16T21_34_04.587307
- 路径:
**/details_harness|drop|3_2023-09-16T21-34-04.587307.parquet - 拆分:latest
- 路径:
**/details_harness|drop|3_2023-09-16T21-34-04.587307.parquet
-
harness_gsm8k_5
- 拆分:2023_09_16T21_34_04.587307
- 路径:
**/details_harness|gsm8k|5_2023-09-16T21-34-04.587307.parquet - 拆分:latest
- 路径:
**/details_harness|gsm8k|5_2023-09-16T21-34-04.587307.parquet
-
harness_hellaswag_10
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hellaswag|10_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hellaswag|10_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T20:43:28.351317.parquet等 39 个文件 - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T20:43:28.351317.parquet等 39 个文件
-
harness_hendrycksTest_abstract_algebra_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_anatomy_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_astronomy_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_business_ethics_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_college_biology_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_college_chemistry_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_college_computer_science_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_college_mathematics_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_college_medicine_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_college_physics_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_computer_security_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_econometrics_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-19T20:43:28.351317.parquet - 拆分:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-19T20:43:28.351317.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 拆分:2023_07_19T20_43_28.351317
- 路径:`**/details_harness|hendrycksTest-electrical_engineering|5_2023



