open-llm-leaderboard-old/details_CalderaAI__13B-Legerdemain-L2
收藏数据集概述
数据集简介
该数据集是在对模型 CalderaAI/13B-Legerdemain-L2 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
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,每次运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳。
- "train" 分割始终指向最新的结果。
- 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。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_CalderaAI__13B-Legerdemain-L2", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-12T20:33:10.328879 运行 的最新结果: python { "all": { "em": 0.002726510067114094, "em_stderr": 0.0005340111700415904, "f1": 0.06216547818791966, "f1_stderr": 0.0013785278979549318, "acc": 0.4412861505062612, "acc_stderr": 0.010705008172209724 }, "harness|drop|3": { "em": 0.002726510067114094, "em_stderr": 0.0005340111700415904, "f1": 0.06216547818791966, "f1_stderr": 0.0013785278979549318 }, "harness|gsm8k|5": { "acc": 0.13040181956027294, "acc_stderr": 0.0092756303245541 }, "harness|winogrande|5": { "acc": 0.7521704814522494, "acc_stderr": 0.01213438601986535 } }
配置详情
以下是数据集的配置详情:
-
harness_arc_challenge_25
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|arc:challenge|25_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-08-09T11:34:37.986977.parquet
-
harness_drop_3
- 分割:2023_10_12T20_33_10.328879
- 路径:
**/details_harness|drop|3_2023-10-12T20-33-10.328879.parquet - 分割:latest
- 路径:
**/details_harness|drop|3_2023-10-12T20-33-10.328879.parquet
-
harness_gsm8k_5
- 分割:2023_10_12T20_33_10.328879
- 路径:
**/details_harness|gsm8k|5_2023-10-12T20-33-10.328879.parquet - 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-12T20-33-10.328879.parquet
-
harness_hellaswag_10
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hellaswag|10_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T11:34:37.986977.parquet等 36 个文件 - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T11:34:37.986977.parquet等 36 个文件
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-09T11:34:37.986977.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-08-09T11:34:37.986977.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_08_09T11_34_37.986977
- 路径:
**/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T11:34:37.986977.parquet



