open-llm-leaderboard-old/details_NousResearch__Nous-Hermes-2-Mixtral-8x7B-DPO
收藏数据集概述
数据集摘要
该数据集是在评估模型 NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO 在 Open LLM Leaderboard 上的评估运行期间自动创建的。
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 2 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_NousResearch__Nous-Hermes-2-Mixtral-8x7B-DPO", "harness_winogrande_5", split="train")
最新结果
这些是最新结果,来自 2024-01-22T17:09:50.643842 的运行: python { "all": { "acc": 0.7224125718980299, "acc_stderr": 0.030022741290236767, "acc_norm": 0.7240829737285515, "acc_norm_stderr": 0.03062607991215834, "mc1": 0.3880048959608323, "mc1_stderr": 0.01705876150134797, "mc2": 0.5482610472622913, "mc2_stderr": 0.014924708991833662 }, "harness|arc:challenge|25": { "acc": 0.6953924914675768, "acc_stderr": 0.013449522109932487, "acc_norm": 0.7107508532423208, "acc_norm_stderr": 0.013250012579393441 }, "harness|hellaswag|10": { "acc": 0.6870145389364668, "acc_stderr": 0.004627607991626919, "acc_norm": 0.8729336785500896, "acc_norm_stderr": 0.0033236659644121946 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.43, "acc_stderr": 0.049756985195624284, "acc_norm": 0.43, "acc_norm_stderr": 0.049756985195624284 }, # 其他任务的结果... }
配置详情
-
config_name: harness_arc_challenge_25
- split: 2024_01_16T04_44_16.630676
- path: **/details_harness|arc:challenge|25_2024-01-16T04-44-16.630676.parquet
- split: 2024_01_22T17_09_50.643842
- path: **/details_harness|arc:challenge|25_2024-01-22T17-09-50.643842.parquet
- split: latest
- path: **/details_harness|arc:challenge|25_2024-01-22T17-09-50.643842.parquet
- split: 2024_01_16T04_44_16.630676
-
config_name: harness_gsm8k_5
- split: 2024_01_16T04_44_16.630676
- path: **/details_harness|gsm8k|5_2024-01-16T04-44-16.630676.parquet
- split: 2024_01_22T17_09_50.643842
- path: **/details_harness|gsm8k|5_2024-01-22T17-09-50.643842.parquet
- split: latest
- path: **/details_harness|gsm8k|5_2024-01-22T17-09-50.643842.parquet
- split: 2024_01_16T04_44_16.630676
-
config_name: harness_hellaswag_10
- split: 2024_01_16T04_44_16.630676
- path: **/details_harness|hellaswag|10_2024-01-16T04-44-16.630676.parquet
- split: 2024_01_22T17_09_50.643842
- path: **/details_harness|hellaswag|10_2024-01-22T17-09-50.643842.parquet
- split: latest
- path: **/details_harness|hellaswag|10_2024-01-22T17-09-50.643842.parquet
- split: 2024_01_16T04_44_16.630676
-
config_name: harness_hendrycksTest_5
- split: 2024_01_16T04_44_16.630676
- path:
- **/details_harness|hendrycksTest-abstract_algebra|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-anatomy|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-astronomy|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-business_ethics|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-clinical_knowledge|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-college_biology|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-college_chemistry|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-college_computer_science|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-college_mathematics|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-college_medicine|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-college_physics|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-computer_security|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-conceptual_physics|5_2024-01-16T04-44-16.630676.parquet
- **/details_harness|hendrycksTest-econometrics|5_2024-01-16T04-44-16.630676.parquet
其他路径...
- path:
- split: 2024_01_16T04_44_16.630676



