open-llm-leaderboard/details_NousResearch__CodeLlama-13b-hf
收藏数据集概述
数据集简介
该数据集是在评估模型 NousResearch/CodeLlama-13b-hf 在 Open LLM Leaderboard 上的运行过程中自动创建的。数据集包含 64 个配置,每个配置对应一个评估任务。
数据集结构
数据集由 2 次运行结果组成,每次运行的结果可以在每个配置中找到,以运行的时间戳命名的特定分片形式存储。"train" 分片始终指向最新的结果。
额外配置
一个额外的配置 "results" 存储了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_NousResearch__CodeLlama-13b-hf", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-18T22:40:09.407812 运行 的最新结果: python { "all": { "em": 0.0010486577181208054, "em_stderr": 0.00033145814652192065, "f1": 0.05248531879194655, "f1_stderr": 0.0012515405190332619, "acc": 0.3964846847094825, "acc_stderr": 0.011095593973496732 }, "harness|drop|3": { "em": 0.0010486577181208054, "em_stderr": 0.00033145814652192065, "f1": 0.05248531879194655, "f1_stderr": 0.0012515405190332619 }, "harness|gsm8k|5": { "acc": 0.12130401819560273, "acc_stderr": 0.008992888497275572 }, "harness|winogrande|5": { "acc": 0.6716653512233622, "acc_stderr": 0.01319829944971789 } }
配置详情
以下是数据集的配置详情:
-
config_name: harness_arc_challenge_25
- split: 2023_08_29T18_13_52.290314
- path: **/details_harness|arc:challenge|25_2023-08-29T18:13:52.290314.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-08-29T18:13:52.290314.parquet
- split: 2023_08_29T18_13_52.290314
-
config_name: harness_drop_3
- split: 2023_10_18T22_40_09.407812
- path: **/details_harness|drop|3_2023-10-18T22-40-09.407812.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-18T22-40-09.407812.parquet
- split: 2023_10_18T22_40_09.407812
-
config_name: harness_gsm8k_5
- split: 2023_10_18T22_40_09.407812
- path: **/details_harness|gsm8k|5_2023-10-18T22-40-09.407812.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-18T22-40-09.407812.parquet
- split: 2023_10_18T22_40_09.407812
-
config_name: harness_hellaswag_10
- split: 2023_08_29T18_13_52.290314
- path: **/details_harness|hellaswag|10_2023-08-29T18:13:52.290314.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-08-29T18:13:52.290314.parquet
- split: 2023_08_29T18_13_52.290314
-
config_name: harness_hendrycksTest_5
- split: 2023_08_29T18_13_52.290314
- path: **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-anatomy|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-astronomy|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-business_ethics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-college_biology|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-college_chemistry|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-college_computer_science|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-college_mathematics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-college_medicine|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-college_physics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-computer_security|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-econometrics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-formal_logic|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-global_facts|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_biology|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_geography|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_physics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-human_aging|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-human_sexuality|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-international_law|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-jurisprudence|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-machine_learning|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-management|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-marketing|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-medical_genetics|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-miscellaneous|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-moral_disputes|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-nutrition|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-philosophy|5_2023-08-29T18:13:52.290314.parquet
- path: **/details_harness|hendrycksTest-prehistory|5_2023-08-29T18:13:52.290314.parquet
- path: **/details
- split: 2023_08_29T18_13_52.290314



