largest_gram = None
    largest_n = 0
    cur_gram = None
    cur_gram_count = 0

    for line in f:
        gram, year, n, books = line.split()

        if cur_gram is None or gram != cur_gram:
            if cur_gram_count > largest_n:
                largest_n = cur_gram_count
                largest_gram = cur_gram        
            cur_gram_count = 0
            cur_gram = gram
        
        cur_gram_count += int(n)

$ sed -e 's/(//' -e 's/)//' -e 's/,//' per-letter
 | awk '{print $2 " " $1}' | sort -nr | head -n 1
23688414489 'the'
$