open-llm-leaderboard-old/details_TehVenom__PPO_Shygmalion-V8p4_Dev-6b
收藏数据集概述
数据集来源
- 数据集自动创建于模型 TehVenom/PPO_Shygmalion-V8p4_Dev-6b 在 Open LLM Leaderboard 的评估运行期间。
数据集结构
- 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。
结果配置
- 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。
最新结果
- 最新结果来自 2023-10-19T06:04:59.904183 运行,包含多个任务的结果。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_TehVenom__PPO_Shygmalion-V8p4_Dev-6b", "harness_winogrande_5", split="train")
配置详情
-
harness_arc_challenge_25
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|arc:challenge|25_2023-07-19T15:59:40.627268.parquet
-
harness_drop_3
- 分割: 2023_10_19T06_04_59.904183, latest
- 路径: **/details_harness|drop|3_2023-10-19T06-04-59.904183.parquet
-
harness_gsm8k_5
- 分割: 2023_10_19T06_04_59.904183, latest
- 路径: **/details_harness|gsm8k|5_2023-10-19T06-04-59.904183.parquet
-
harness_hellaswag_10
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hellaswag|10_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: 包含多个学科的 parquet 文件路径
-
harness_hendrycksTest_abstract_algebra_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_anatomy_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-anatomy|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_astronomy_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-astronomy|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_college_biology_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-college_biology|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_college_physics_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-college_physics|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_computer_security_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-computer_security|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_econometrics_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-econometrics|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T15:59:40.627268.parquet
-
harness_hendrycksTest_elementary_mathematics_5
- 分割: 2023_07_19T15_59_40.627268, latest
- 路径: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T15:59:40.627268.parquet



