its not going to work blocking the urls or directories which do not present on your server. If they are not there, which you are going to block? The googlebot is picking it from somewhere else and not from your root directory. You can control your own server but not of others. You may have submitted an article that contains those urls as anchor link (of course by article directories' programming error), now you need to find out the source and correct the links to point to the valid url. In the meantime you can do a 301 redirect to some page on your website to avoid 404 error.