open-llm-leaderboard/details_CHIH-HUNG__llama-2-13b-FINETUNE5_4w-r8-q_k_v_o_gate_up_down
收藏数据集概述
数据集来源
该数据集是在评估模型 CHIH-HUNG/llama-2-13b-FINETUNE5_4w-r8-q_k_v_o_gate_up_down 时自动创建的,用于 Open LLM Leaderboard。
数据集结构
- 配置数量:64个配置,每个配置对应一个评估任务。
- 数据来源:数据集由2次运行创建,每次运行可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。
- 最新结果:"train" 分割始终指向最新结果。
- 汇总结果:额外配置 "results" 存储所有运行结果的汇总,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_CHIH-HUNG__llama-2-13b-FINETUNE5_4w-r8-q_k_v_o_gate_up_down", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-25T21:40:15.944875 运行 的最新结果: python { "all": { "em": 0.07697147651006711, "em_stderr": 0.002729682408788614, "f1": 0.12191170302013389, "f1_stderr": 0.0028589398116221384, "acc": 0.44546584943629414, "acc_stderr": 0.01035635936441261 }, "harness|drop|3": { "em": 0.07697147651006711, "em_stderr": 0.002729682408788614, "f1": 0.12191170302013389, "f1_stderr": 0.0028589398116221384 }, "harness|gsm8k|5": { "acc": 0.11902956785443518, "acc_stderr": 0.00891970291116163 }, "harness|winogrande|5": { "acc": 0.7719021310181531, "acc_stderr": 0.011793015817663592 } }
配置详情
-
harness_arc_challenge_25
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|arc:challenge|25_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|arc:challenge|25_2023-10-10T10-20-42.158103.parquet
-
harness_drop_3
- 分割:2023_10_25T21_40_15.944875
- 路径:
**/details_harness|drop|3_2023-10-25T21-40-15.944875.parquet - 分割:latest
- 路径:
**/details_harness|drop|3_2023-10-25T21-40-15.944875.parquet
-
harness_gsm8k_5
- 分割:2023_10_25T21_40_15.944875
- 路径:
**/details_harness|gsm8k|5_2023-10-25T21-40-15.944875.parquet - 分割:latest
- 路径:
**/details_harness|gsm8k|5_2023-10-25T21-40-15.944875.parquet
-
harness_hellaswag_10
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hellaswag|10_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hellaswag|10_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T10-20-42.158103.parquet等36个文件 - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T10-20-42.158103.parquet等36个文件
-
harness_hendrycksTest_abstract_algebra_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-abstract_algebra|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_anatomy_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-anatomy|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_astronomy_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-astronomy|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_business_ethics_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-business_ethics|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_clinical_knowledge_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-clinical_knowledge|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_college_biology_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_biology|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_college_chemistry_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_chemistry|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_college_computer_science_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_computer_science|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_college_mathematics_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_mathematics|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_college_medicine_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_medicine|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_college_physics_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-college_physics|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_computer_security_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-computer_security|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_conceptual_physics_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-conceptual_physics|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_econometrics_5
- 分割:2023_10_10T10_20_42.158103
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-10T10-20-42.158103.parquet - 分割:latest
- 路径:
**/details_harness|hendrycksTest-econometrics|5_2023-10-10T10-20-42.158103.parquet
-
harness_hendrycksTest_electrical_engineering_5



