open-llm-leaderboard-old/details_togethercomputer__RedPajama-INCITE-Chat-7B-v0.1
收藏数据集概述
该数据集是在对模型 togethercomputer/RedPajama-INCITE-Chat-7B-v0.1 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
数据集组成
数据集由 64 个配置组成,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找到特定的分割,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 Open LLM Leaderboard 上的聚合指标。
加载数据示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_togethercomputer__RedPajama-INCITE-Chat-7B-v0.1", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-19T01:29:17.433845 运行的最新结果:
python { "all": { "em": 0.00985738255033557, "em_stderr": 0.001011740962658439, "f1": 0.06564072986577182, "f1_stderr": 0.0016570971110147965, "acc": 0.3014062577602678, "acc_stderr": 0.007815997155326552 }, "harness|drop|3": { "em": 0.00985738255033557, "em_stderr": 0.001011740962658439, "f1": 0.06564072986577182, "f1_stderr": 0.0016570971110147965 }, "harness|gsm8k|5": { "acc": 0.004548900682335102, "acc_stderr": 0.0018535550440036204 }, "harness|winogrande|5": { "acc": 0.5982636148382005, "acc_stderr": 0.013778439266649482 } }
配置详情
-
harness_arc_challenge_25
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|arc:challenge|25_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_drop_3
- 分割:
2023_10_19T01_29_17.433845,latest - 路径:
**/details_harness|drop|3_2023-10-19T01-29-17.433845.parquet
- 分割:
-
harness_gsm8k_5
- 分割:
2023_10_19T01_29_17.433845,latest - 路径:
**/details_harness|gsm8k|5_2023-10-19T01-29-17.433845.parquet
- 分割:
-
harness_hellaswag_10
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hellaswag|10_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径: 多个路径,例如
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_abstract_algebra_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_anatomy_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_astronomy_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_business_ethics_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_college_biology_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_college_chemistry_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_college_computer_science_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_college_mathematics_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_college_medicine_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_college_physics_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_computer_security_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_conceptual_physics_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_econometrics_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-19T16:36:55.305122.parquet
- 分割:
-
harness_hendrycksTest_electrical_engineering_5
- 分割:
2023_07_19T16_36_55.305122,latest - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T16:36:55.305122.parquet
- 分割:
以上是数据集的详细概述,包括数据集的组成、加载示例、最新结果以及各个配置的详细信息。



