open-llm-leaderboard/details_togethercomputer__LLaMA-2-7B-32K
收藏数据集概述
数据集简介
该数据集是在对模型 togethercomputer/LLaMA-2-7B-32K 进行评估运行期间自动创建的,用于在 Open LLM Leaderboard 上展示评估结果。
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 3 次运行中创建,每个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" 分割总是指向最新的结果。
- 一个额外的配置 "results" 存储所有运行的聚合结果,用于计算和显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_togethercomputer__LLaMA-2-7B-32K", "harness_winogrande_5", split="train")
最新结果
以下是 2023-10-13T06:14:11.613240 运行的最新结果: python { "all": { "em": 0.001572986577181208, "em_stderr": 0.0004058451132417743, "f1": 0.05483536073825504, "f1_stderr": 0.0013143577122089422, "acc": 0.3811179333318573, "acc_stderr": 0.009116764305556838 }, "harness|drop|3": { "em": 0.001572986577181208, "em_stderr": 0.0004058451132417743, "f1": 0.05483536073825504, "f1_stderr": 0.0013143577122089422 }, "harness|gsm8k|5": { "acc": 0.043214556482183475, "acc_stderr": 0.0056009875152378515 }, "harness|winogrande|5": { "acc": 0.7190213101815311, "acc_stderr": 0.012632541095875824 } }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2023_08_09T09_44_03.510382
- path: **/details_harness|arc:challenge|25_2023-08-09T09:44:03.510382.parquet
- split: 2023_08_09T14_19_55.056276
- path: **/details_harness|arc:challenge|25_2023-08-09T14:19:55.056276.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-08-09T14:19:55.056276.parquet
- split: 2023_08_09T09_44_03.510382
-
config_name: harness_drop_3
- split: 2023_10_13T06_14_11.613240
- path: **/details_harness|drop|3_2023-10-13T06-14-11.613240.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-13T06-14-11.613240.parquet
- split: 2023_10_13T06_14_11.613240
-
config_name: harness_gsm8k_5
- split: 2023_10_13T06_14_11.613240
- path: **/details_harness|gsm8k|5_2023-10-13T06-14-11.613240.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-13T06-14-11.613240.parquet
- split: 2023_10_13T06_14_11.613240
-
config_name: harness_hellaswag_10
- split: 2023_08_09T09_44_03.510382
- path: **/details_harness|hellaswag|10_2023-08-09T09:44:03.510382.parquet
- split: 2023_08_09T14_19_55.056276
- path: **/details_harness|hellaswag|10_2023-08-09T14:19:55.056276.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-08-09T14:19:55.056276.parquet
- split: 2023_08_09T09_44_03.510382
-
config_name: harness_hendrycksTest_5
- split: 2023_08_09T09_44_03.510382
- path: **/details_harness|hendrycksTest-abstract_algebra|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-anatomy|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-astronomy|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-business_ethics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-clinical_knowledge|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-college_biology|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-college_chemistry|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-college_computer_science|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-college_mathematics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-college_medicine|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-college_physics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-computer_security|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-conceptual_physics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-econometrics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-electrical_engineering|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-elementary_mathematics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-formal_logic|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-global_facts|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_biology|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_chemistry|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_computer_science|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_european_history|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_geography|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_mathematics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_physics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_psychology|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_statistics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_us_history|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-high_school_world_history|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-human_aging|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-human_sexuality|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-international_law|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-jurisprudence|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-logical_fallacies|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-machine_learning|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-management|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-marketing|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-medical_genetics|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-miscellaneous|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-moral_disputes|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-moral_scenarios|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_harness|hendrycksTest-nutrition|5_2023-08-09T09:44:03.510382.parquet
- path: **/details_
- split: 2023_08_09T09_44_03.510382



