Expected? json data generated from convert conll-u shows unicode?

[yosiasz]

hello,

I am converting amharic conllu file using the following command

spacy convert --merge-subtokens '.\corpus\conllu\am_att-ud-test.conllu' -c conllu ./output/ -t json

and it generates this

{
                "id":1,
                "orth":"\u12a0\u12ed\u1296\u1279",
                "space":" ",
                "tag":"NOUN_POSM",
                "pos":"NOUN",
                "morph":"Gender=Masc|Number=Plur,Sing|Person=3|Poss=Yes|PronType=Prs",
                "lemma":"\u12a0\u12ed\u1296\u127d \u12a5\u1231",
                "head":2,
                "dep":"nsubj"
              },

See orth and lemma

It is not showing me actual Amharic words but uxyz is that to be expected or is this a bug in spaCy? Hard to read this unless I use another external tool to parse these?

Thanks!