You can find all files that are probably duplicates using the following command:
find -type f -exec md5sum '{}' ';' | sort | uniq --all-repeated=separate -w 33 | cut -c 35-
They can then be fully compared to each other using diff
diff file1 file2