Note however that while the Dictionary was 5,000 lemmas, on Lextutor you find the complete 25 k-list analysis from which the dictionary lists were taken (generously contributed by Lonsdale on the understanding they would be used for research purposes).
Flesh-out of the raw lemma list was accomplished at Lextutor in summer 2013 with lemma resources from lexique.org. The Lonsdale and Lebras lists are based on a dedicated 23-million-word corpus of French which includes a balanced sample of both written and spoken material, both literary and non-literary material, from both France and other places (mainly Canada) where French is spoken, and employs criteria of both frequency and range (distribution throughout the corpus rather than just in one part of it). The corpus compositon summary is as follows:
It is important to note that these word lists employ the lemma as the highest level unit rather than the family, which is used in the Laufer-Nation-group VPs. Lemmas are base words plus inflections (chat chats; viens vient) that do not alter the part of speech; families are also lemmas, but in addition include "obvious derivations" that do change the part of speech (saluer, saluez, etc but also salut).
A families version of this list is now available (Fams 1-3k) including nuclear versions from NFL-x links (with related article in preparation late Summer 2023)