Surprisingly there is no accepted answer. The issue only exists in 32-bit PHP.
From the documentation,
If the string does not contain any of the characters '.', 'e', or 'E' and the numeric value fits into integer type limits (as defined by PHP_INT_MAX), the string will be evaluated as an integer. In all other cases it will be evaluated as a float.
In other words, the $string is first interpreted as INT, which cause overflow (The $string value 2968789218 exceeds the maximum value (PHP_INT_MAX) of 32-bit PHP, which is 2147483647.), then evaluated to float by (float) or floatval().
Thus, the solution is:
$string = "2968789218";
echo 'Original: ' . floatval($string) . PHP_EOL;
$string.= ".0";
$float = floatval($string);
echo 'Corrected: ' . $float . PHP_EOL;
which outputs:
Original: 2.00
Corrected: 2968789218
To check whether your PHP is 32-bit or 64-bit, you can:
echo PHP_INT_MAX;
If your PHP is 64-bit, it will print out 9223372036854775807, otherwise it will print out 2147483647.