Using a Genetic algorithm to evolve to a specified string

Question

so im working on a genetic algorithm that is meant to converge on a specific string, this string is a flattened asccii art, im using a sequenceMatcher to find the ratio of how similar a given chromosone is to the answer as my fitness, the code is as below:

import random
from difflib import SequenceMatcher

GA="0001111000001100000100001000101100010000000010001100010011100111111000011101001000010"
print(len(GA))


# Hyper-parameter
no_pop = 10 # population size
no_sel = 5   # slect the best no_sel idnividuals
no_gen = 10
p_xo = .8  # crossover probabilty/rate
p_m = 0.05 # mutation probabilty/rate per bit


# 1 Step: Select representation
no_bit = 85 # number of bits



# Create the pop
pop = [] # population
for i in range(no_pop):
    pop.append(''.join(random.choice('01') for j in range(no_bit)))
    
pop

# convert bitstring into decimal number
def dec(chromosome):
    d = 0
    for i in range(len(chromosome)):
        if chromosome[i] == '1':
            d += 2**i
    return d

def fitness(chromosome):
    return SequenceMatcher(None,str(dec(chromosome)),str(dec(GA))).ratio()  #check similarity to answer

def one_point_crossover(chromosome_1, chromosome_2):
    cut = random.randint(1, no_bit-1) # randomy select cut point
    offspring = chromosome_1[0:cut] + chromosome_2[cut:]
    return offspring


def bit_flip_mutation(chromosome):
    mutation = ''
    for i in range(len(chromosome)):
        if random.random() < p_m:
            if chromosome[i] == '0':
                mutation += '1'
            else:
                mutation += '0'
        else:
            mutation += chromosome[i]

    return chromosome

for g in range(no_gen):
    fit = []
    for i in range(no_pop):
        fit.append(fitness(pop[i]))
        
    print('Gen' + str(g))
    print(pop)
    print(fit)
    
    index = sorted(range(len(fit)), key=lambda k: fit[k], reverse=True)
    
    pop_new = []
    for i in index[:no_sel]:
        pop_new.append(pop[i])
        
    for i in range(no_sel, no_pop):
        parent_1 = random.randint(0, no_sel-1)
        parent_2 = random.randint(0, no_sel-1)
        offspring = one_point_crossover(pop_new[parent_1], pop_new[parent_2])
        offspring = bit_flip_mutation(offspring)
        pop_new.append(offspring)
        
    pop = pop_new
   
fit = []
for i in range(no_pop):
    fit.append(fitness(pop[i]))

print('Gen' + str(g))
print(pop)
print(fit)

currently the algorthim doesnt achieve mpore than around 57% simmilarity to the actual answer. im confused as to why this is, any siggesttions on how to improve this would be great.