open-llm-leaderboard-old/details_MaziyarPanahi__UNA-34Beagles-32K-bf16-v1-GPTQ
收藏数据集概述
该数据集是在对模型 MaziyarPanahi/UNA-34Beagles-32K-bf16-v1-GPTQ 进行评估运行期间自动创建的,用于 Open LLM Leaderboard。
数据集组成
- 数据集包含 63 个配置,每个配置对应一个评估任务。
- 数据集从 1 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。
- 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_MaziyarPanahi__UNA-34Beagles-32K-bf16-v1-GPTQ", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的摘要:
python { "all": { "acc": 0.24429336199406124, "acc_stderr": 0.030450678156342035, "acc_norm": 0.24486601555293414, "acc_norm_stderr": 0.03125930165483659, "mc1": 0.22888616891064872, "mc1_stderr": 0.014706994909055027, "mc2": 0.47265741103126685, "mc2_stderr": 0.01706712892538534 }, "harness|arc:challenge|25": { "acc": 0.2158703071672355, "acc_stderr": 0.012022975360030668, "acc_norm": 0.26109215017064846, "acc_norm_stderr": 0.01283552390947385 }, "harness|hellaswag|10": { "acc": 0.2546305516829317, "acc_stderr": 0.004347629889040941, "acc_norm": 0.26289583748257317, "acc_norm_stderr": 0.004393066760916824 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.3, "acc_stderr": 0.046056618647183814, "acc_norm": 0.3, "acc_norm_stderr": 0.046056618647183814 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2024_02_19T03_20_06.212464
- path:
**/details_harness|arc:challenge|25_2024-02-19T03-20-06.212464.parquet
- path:
- split: latest
- path:
**/details_harness|arc:challenge|25_2024-02-19T03-20-06.212464.parquet
- path:
- split: 2024_02_19T03_20_06.212464
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2024_02_19T03_20_06.212464
- path:
**/details_harness|gsm8k|5_2024-02-19T03-20-06.212464.parquet
- path:
- split: latest
- path:
**/details_harness|gsm8k|5_2024-02-19T03-20-06.212464.parquet
- path:
- split: 2024_02_19T03_20_06.212464
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2024_02_19T03_20_06.212464
- path:
**/details_harness|hellaswag|10_2024-02-19T03-20-06.212464.parquet
- path:
- split: latest
- path:
**/details_harness|hellaswag|10_2024-02-19T03-20-06.212464.parquet
- path:
- split: 2024_02_19T03_20_06.212464
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2024_02_19T03_20_06.212464
- path:
**/details_harness|hendrycksTest-abstract_algebra|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-anatomy|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-astronomy|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-business_ethics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-clinical_knowledge|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-college_biology|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-college_chemistry|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-college_computer_science|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-college_mathematics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-college_medicine|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-college_physics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-computer_security|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-conceptual_physics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-econometrics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-electrical_engineering|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-elementary_mathematics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-formal_logic|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-global_facts|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_biology|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_chemistry|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_computer_science|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_european_history|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_geography|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_government_and_politics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_macroeconomics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_mathematics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_microeconomics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_physics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_psychology|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_statistics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_us_history|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-high_school_world_history|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-human_aging|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-human_sexuality|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-international_law|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-jurisprudence|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-logical_fallacies|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-machine_learning|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-management|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-marketing|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-medical_genetics|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-miscellaneous|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-moral_disputes|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-moral_scenarios|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-nutrition|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-philosophy|5_2024-02-19T03-20-06.212464.parquet**/details_harness|hendrycksTest-prehistory|5_2024-02-19T03-20-06.212464.parquet- `**/details_harness|hendrycksTest-
- path:
- split: 2024_02_19T03_20_06.212464
- data_files:



