open-llm-leaderboard/details_notstoic__PygmalionCoT-7b
收藏数据集概述
该数据集是在对模型 notstoic/PygmalionCoT-7b 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 64 个配置,每个配置对应一个评估任务。
- 数据集从 2 次运行中创建,每次运行可以在每个配置中作为一个特定的分片找到,分片名称使用运行的时间戳。
- "train" 分片始终指向最新的结果。
- 一个额外的配置 "results" 存储所有运行结果的聚合,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_notstoic__PygmalionCoT-7b", "harness_winogrande_5", split="train")
最新结果
以下是 2023-09-22T15:06:38.792335 运行的最新结果:
python { "all": { "em": 0.12111996644295302, "em_stderr": 0.0033412757702121106, "f1": 0.17514471476510068, "f1_stderr": 0.0034689450739406216, "acc": 0.36081482886571287, "acc_stderr": 0.00895060187911282 }, "harness|drop|3": { "em": 0.12111996644295302, "em_stderr": 0.0033412757702121106, "f1": 0.17514471476510068, "f1_stderr": 0.0034689450739406216 }, "harness|gsm8k|5": { "acc": 0.032600454890068235, "acc_stderr": 0.004891669021939579 }, "harness|winogrande|5": { "acc": 0.6890292028413575, "acc_stderr": 0.01300953473628606 } }
配置详情
-
harness_arc_challenge_25
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|arc:challenge|25_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|arc:challenge|25_2023-07-18T12:24:33.017908.parquet
-
harness_drop_3
- 分片:2023_09_22T15_06_38.792335
- 路径:
**/details_harness|drop|3_2023-09-22T15-06-38.792335.parquet - 分片:latest
- 路径:
**/details_harness|drop|3_2023-09-22T15-06-38.792335.parquet
-
harness_gsm8k_5
- 分片:2023_09_22T15_06_38.792335
- 路径:
**/details_harness|gsm8k|5_2023-09-22T15-06-38.792335.parquet - 分片:latest
- 路径:
**/details_harness|gsm8k|5_2023-09-22T15-06-38.792335.parquet
-
harness_hellaswag_10
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hellaswag|10_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hellaswag|10_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T12:24:33.017908.parquet等 40 个文件 - 分片:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T12:24:33.017908.parquet等 40 个文件
-
harness_hendrycksTest_abstract_algebra_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_anatomy_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_astronomy_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_business_ethics_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_college_biology_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_college_medicine_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_college_physics_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_computer_security_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_econometrics_5
- 分片:2023_07_18T12_24_33.017908
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-18T12:24:33.017908.parquet - 分片:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-18T12:24:33.017908.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分片:2023_07_18T12_24_33.017908
- 路径:`**/details_harness|hendrycksTest-electrical_engineering|5_2023



