Define the example in the question and call it `a`

:

```
sage: a = CoxeterGroup(['A20']).gens()[0]
```

It displays as

```
sage: a
[-1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1]
```

Define the same matrix starting from the identity matrix
and modifying the first two entries; call that `b`

:

```
sage: b = identity_matrix(20)
sage: b[0:2, 0:0] = (-1, 1)
```

Check that `a`

and `b`

define the same matrix:

```
sage: b == a
True
```

However `b`

displays differently:

```
sage: b
20 x 20 dense matrix over Integer Ring
(use the '.str()' method to see the entries)
```

Get the fuller display of `b`

, showing all its entries, using `print`

:

```
sage: print(b)
[-1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1]
```

Let us explain this difference in display,
and how to get the summary display for `a`

.

Even though `a`

and `b`

have the same entries, they don't have
the same "parent"; `a`

is an element in a matrix group while
`b`

is an element in a matrix space.

```
sage: a.parent()
Finite Coxeter group over Integer Ring with Coxeter matrix:
20 x 20 dense matrix over Integer Ring
sage: b.parent()
Full MatrixSpace of 20 by 20 dense matrices over Integer Ring
```

The difference in display comes from a different implementation
of the `__repr__`

method for elements of matrix groups and elements
of matrix spaces: compare `a.__repr__??`

and `b.__repr__??`

.

To get the summary version for `a`

, one solution is to convert
it to a matrix space element.

```
sage: c = matrix(a)
sage: c
20 x 20 dense matrix over Integer Ring
(use the '.str()' method to see the entries)
```

To get elements of matrix groups to consistently use the
shorter display when they reach 20 rows, one would have
to change the `__repr__`

method of matrix group elements.

Could you include a very concrete example of the desired input and output?

Thanks for your edit, adding an example. It helps understanding and answering the question.

Please also fix the "curly quotes" (maybe "smart quotes" is turned on in your operating system).

Copy pasting the current

results in an error, while with straight quotes it works: