open-llm-leaderboard/details_stabilityai__stablelm-tuned-alpha-3b
收藏数据集概述
该数据集是在评估模型 stabilityai/stablelm-tuned-alpha-3b 的过程中自动创建的,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" 存储了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_stabilityai__stablelm-tuned-alpha-3b", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-15T21:41:17.455218 运行 的最新结果:
python { "all": { "em": 0.0036703020134228187, "em_stderr": 0.0006192871806511154, "f1": 0.05061136744966456, "f1_stderr": 0.00134828623344778, "acc": 0.27771272034672656, "acc_stderr": 0.007991508812498901 }, "harness|drop|3": { "em": 0.0036703020134228187, "em_stderr": 0.0006192871806511154, "f1": 0.05061136744966456, "f1_stderr": 0.00134828623344778 }, "harness|gsm8k|5": { "acc": 0.00530705079605762, "acc_stderr": 0.002001305720948078 }, "harness|winogrande|5": { "acc": 0.5501183898973955, "acc_stderr": 0.013981711904049725 } }
配置详情
-
harness_arc_challenge_25
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|arc:challenge|25_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_drop_3
- 分割:
2023_10_15T21_41_17.455218,latest - 路径:
**/details_harness|drop|3_2023-10-15T21-41-17.455218.parquet
- 分割:
-
harness_gsm8k_5
- 分割:
2023_10_15T21_41_17.455218,latest - 路径:
**/details_harness|gsm8k|5_2023-10-15T21-41-17.455218.parquet
- 分割:
-
harness_hellaswag_10
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hellaswag|10_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径: 多个路径,例如
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_abstract_algebra_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_anatomy_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_astronomy_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_business_ethics_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_college_biology_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_college_chemistry_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_college_computer_science_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_college_mathematics_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_college_medicine_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_college_physics_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_computer_security_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_conceptual_physics_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_econometrics_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_electrical_engineering_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T14:49:37.876156.parquet
- 分割:
-
harness_hendrycksTest_elementary_mathematics_5
- 分割:
2023_07_19T14_49_37.876156,latest - 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T14:49:37.876156.parquet
- 分割:



