关于机器学习中岭回归的参数问题-负的均方误差结果为正值问题

各位前辈好，本人最近在学习机器学习的过程中遇到如下问题：本人选用两组随机数作为特征和标签，目的是验证岭回归对随机选择的数据拟合效果差。但在此过程中，发现使用"neg_mean_squared_error"负的均方误差作为评分指标时，结果出现了正值，希望得到前辈的指点，谢谢！

import numpy as np
import pandas as pd
from sklearn.linear_model import RidgeCV
 
rng = np.random.RandomState(42)
X = rng.randn(100, 80)
y = rng.randn(100)
Ridge_ = RidgeCV(alphas=np.arange(1,1001,100)
                 ,scoring="neg_mean_squared_error"
                 ,store_cv_values=True
                 #,cv=5
                ).fit(X, y)
 
[*zip(np.arange(1,1001,100),Ridge_.cv_values_.mean(axis=0))]

[(1, -0.017030206350983942),

(101, -0.0006180241231415761),
(201, 0.0008785369000308918),
(301, 0.0011539924905547415),
(401, 0.0011727715611324074),
(501, 0.0011223783355600165),
(601, 0.0010543033304316618),
(701, 0.000985010448300175),
(801, 0.0009199279049377978),
(901, 0.0008605959504101133)]