how to the select the most informative columns in a dataframe?

Question

I want to select the most representation columns in a dataFrame. I read somewhere that: " The eigenvectors represent the principal components that contain most of the information (variance) represented using features (independent variables)." So, I do like that:

import numpy as np
from numpy.linalg import eigh
cov_matrix = np.cov(d, rowvar=False)
#
# Determine eigenvalues and eigenvectors
#
egnvalues, egnvectors = eigh(cov_matrix)
#
# Determine explained variance
#
total_egnvalues = sum(egnvalues)

var_exp = [(i/total_egnvalues) for i in sorted(egnvalues, reverse=True)]
print(total_egnvalues)
print(var_exp)

output:

0.11556921475095604
[0.3839056474139497, 0.22463482813643307, 0.1159777232847966, ...]

So, please correct me (according to these results which columns represent the top2 or top3 columns to consider?