open-llm-leaderboard-old/details_OpenAssistant__stablelm-7b-sft-v7-epoch-3
收藏数据集概述
数据集名称
Evaluation run of OpenAssistant/stablelm-7b-sft-v7-epoch-3
数据集描述
该数据集是在评估模型 OpenAssistant/stablelm-7b-sft-v7-epoch-3 在 Open LLM Leaderboard 上的运行过程中自动创建的。
数据集组成
数据集由64个配置组成,每个配置对应一个评估任务。数据集从3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。
额外配置
一个额外的配置 "results" 存储所有运行的聚合结果,用于计算并在 Open LLM Leaderboard 上显示聚合指标。
数据加载示例
python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_OpenAssistant__stablelm-7b-sft-v7-epoch-3", "harness_winogrande_5", split="train")
最新结果
以下是 最新结果 的示例: python { "all": { "em": 0.05578859060402685, "em_stderr": 0.0023504280872280073, "f1": 0.10613569630872476, "f1_stderr": 0.0026144580255279513, "acc": 0.27616530425036784, "acc_stderr": 0.007839405520583978 }, "harness|drop|3": { "em": 0.05578859060402685, "em_stderr": 0.0023504280872280073, "f1": 0.10613569630872476, "f1_stderr": 0.0026144580255279513 }, "harness|gsm8k|5": { "acc": 0.0037907505686125853, "acc_stderr": 0.0016927007401501943 }, "harness|winogrande|5": { "acc": 0.5485398579321231, "acc_stderr": 0.01398611030101776 } }
配置详情
-
config_name: harness_arc_challenge_25
- data_files:
- split: 2023_07_19T17_06_42.731727
- path: **/details_harness|arc:challenge|25_2023-07-19T17:06:42.731727.parquet
- split: 2023_07_19T17_07_54.588127
- path: **/details_harness|arc:challenge|25_2023-07-19T17:07:54.588127.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2023-07-19T17:07:54.588127.parquet
- split: 2023_07_19T17_06_42.731727
- data_files:
-
config_name: harness_drop_3
- data_files:
- split: 2023_10_13T03_23_25.661445
- path: **/details_harness|drop|3_2023-10-13T03-23-25.661445.parquet
- split: latest
- path: **/details_harness|drop|3_2023-10-13T03-23-25.661445.parquet
- split: 2023_10_13T03_23_25.661445
- data_files:
-
config_name: harness_gsm8k_5
- data_files:
- split: 2023_10_13T03_23_25.661445
- path: **/details_harness|gsm8k|5_2023-10-13T03-23-25.661445.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2023-10-13T03-23-25.661445.parquet
- split: 2023_10_13T03_23_25.661445
- data_files:
-
config_name: harness_hellaswag_10
- data_files:
- split: 2023_07_19T17_06_42.731727
- path: **/details_harness|hellaswag|10_2023-07-19T17:06:42.731727.parquet
- split: 2023_07_19T17_07_54.588127
- path: **/details_harness|hellaswag|10_2023-07-19T17:07:54.588127.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2023-07-19T17:07:54.588127.parquet
- split: 2023_07_19T17_06_42.731727
- data_files:
-
config_name: harness_hendrycksTest_5
- data_files:
- split: 2023_07_19T17_06_42.731727
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-anatomy|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-astronomy|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-college_biology|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-college_physics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-computer_security|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-econometrics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-formal_logic|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-global_facts|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_biology|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_geography|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_physics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-human_aging|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-human_sexuality|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-international_law|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-jurisprudence|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-machine_learning|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-management|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-marketing|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-medical_genetics|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-miscellaneous|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-moral_disputes|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-19T17:06:42.731727.parquet
- **/details_harness|hendrycksTest-nutrition|5_2023-
- path:
- split: 2023_07_19T17_06_42.731727
- data_files:



