Reading here: https://dumps.wikimedia.org/other/enterprise_html/
It says:
Each dump output file consists of a tar.gz archive which, when uncompressed and untarred, contains one file
Emphasis mine. That is not true. Tar contains multiple files. E.g.:
enwiki_0.ndjson enwiki_10.ndjson enwiki_11.ndjson enwiki_12.ndjson enwiki_13.ndjson enwiki_14.ndjson enwiki_15.ndjson enwiki_16.ndjson enwiki_17.ndjson enwiki_18.ndjson enwiki_19.ndjson enwiki_1.ndjson enwiki_20.ndjson enwiki_21.ndjson enwiki_22.ndjson enwiki_23.ndjson enwiki_24.ndjson enwiki_25.ndjson enwiki_26.ndjson enwiki_27.ndjson enwiki_28.ndjson enwiki_2.ndjson enwiki_3.ndjson enwiki_4.ndjson enwiki_5.ndjson enwiki_6.ndjson enwiki_7.ndjson enwiki_8.ndjson enwiki_9.ndjson