open-llm-leaderboard-old/details_CHIH-HUNG__llama-2-13b-FINETUNE3_3.3w-r8-q_k_v_o_gate_up_down
收藏数据集概述
数据集简介
该数据集是在对模型 CHIH-HUNG/llama-2-13b-FINETUNE3_3.3w-r8-q_k_v_o_gate_up_down 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。
- 数据集由 2 次运行创建,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。
- 额外的配置 "results" 存储所有运行结果的聚合,用于计算和显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_CHIH-HUNG__llama-2-13b-FINETUNE3_3.3w-r8-q_k_v_o_gate_up_down", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-28T08:25:23.619577 运行的最新结果: python { "all": { "em": 0.3410234899328859, "em_stderr": 0.0048547488279167, "f1": 0.3798521392617453, "f1_stderr": 0.004781623142650588, "acc": 0.43779100069232807, "acc_stderr": 0.010221026790242026 }, "harness|drop|3": { "em": 0.3410234899328859, "em_stderr": 0.0048547488279167, "f1": 0.3798521392617453, "f1_stderr": 0.004781623142650588 }, "harness|gsm8k|5": { "acc": 0.10841546626231995, "acc_stderr": 0.008563852506627504 }, "harness|winogrande|5": { "acc": 0.7671665351223362, "acc_stderr": 0.011878201073856546 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|arc:challenge|25_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-10-04T05-37-11.185661.parquet
-
harness_drop_3
- 分割:2023_10_28T08_25_23.619577
- 路径:
**/details_harness|drop|3_2023-10-28T08-25-23.619577.parquet - 分割:latest
- 路径:
**/details_harness|drop|3_2023-10-28T08-25-23.619577.parquet
-
harness_gsm8k_5
- 分割:2023_10_28T08_25_23.619577
- 路径:
**/details_harness|gsm8k|5_2023-10-28T08-25-23.619577.parquet - 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-28T08-25-23.619577.parquet
-
harness_hellaswag_10
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hellaswag|10_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T05-37-11.185661.parquet等 50 个文件 - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T05-37-11.185661.parquet等 50 个文件
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_10_04T05_37_11.185661
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-04T05-37-11.185661.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-04T05-37-11.185661.parquet
-
harness_hendrycksTest_electrical_engineering_5
- 分割:2023_10_04T05_37_11.185661
- 路径:`



