open-llm-leaderboard/details_Writer__camel-5b-hf
收藏数据集概述
数据集名称
- Evaluation run of Writer/camel-5b-hf
数据集来源
- 该数据集是在对模型 Writer/camel-5b-hf 进行评估运行期间自动创建的。
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" 存储所有运行的聚合结果,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_Writer__camel-5b-hf", "harness_winogrande_5", split="train")
最新结果
- 这些是最新结果,来自 2023-10-18T14:36:32.116490 的运行: python { "all": { "em": 0.08294882550335571, "em_stderr": 0.0028244998601496944, "f1": 0.14997168624161072, "f1_stderr": 0.003145718068946184, "acc": 0.3069466775731776, "acc_stderr": 0.007700124028579334 }, "harness|drop|3": { "em": 0.08294882550335571, "em_stderr": 0.0028244998601496944, "f1": 0.14997168624161072, "f1_stderr": 0.003145718068946184 }, "harness|gsm8k|5": { "acc": 0.0037907505686125853, "acc_stderr": 0.0016927007401502051 }, "harness|winogrande|5": { "acc": 0.6101026045777427, "acc_stderr": 0.013707547317008463 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|arc:challenge|25_2023-07-19T15:25:02.904083.parquet
-
harness_drop_3
- 分割:2023_10_18T14_36_32.116490, latest
- 路径:
**/details_harness|drop|3_2023-10-18T14-36-32.116490.parquet
-
harness_gsm8k_5
- 分割:2023_10_18T14_36_32.116490, latest
- 路径:
**/details_harness|gsm8k|5_2023-10-18T14-36-32.116490.parquet
-
harness_hellaswag_10
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hellaswag|10_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:多个路径,例如
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T15:25:02.904083.parquet
-
harness_hendrycksTest_elementary_mathematics_5
- 分割:2023_07_19T15_25_02.904083, latest
- 路径:
**/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T15:25:02.904083.parquet
以上是数据集的详细概述,包括数据集的名称、来源、组成、加载示例、最新结果以及各个配置的详细信息。



