Table 4 Statistical values on the split of non-toxic dialogues by interlocutor for each category predicted by MS Content Moderator.

From: A dataset of synthetic art dialogues with ChatGPT

Non-Toxic Split

Expert

User

mean

std

min

max

mean

std

min

max

category1

0.0104

0.0401

0

0.875

0.0086

0.0188

0

0.747

category2

0.1172

0.0751

0.001

0.997

0.1719

0.0544

0.001

0.744

category3

0.1501

0.1976

0.022

0.988

0.1354

0.1169

0.022

0.988

Back to article page