INDEX
Explanations
themes related to mental health, emotional distress, and supportive relationships
New Auto-Interp
Negative Logits
arges
-0.17
ugin
-0.15
emen
-0.15
illon
-0.14
bare
-0.14
resh
-0.14
ong
-0.14
غÙĨ
-0.14
£
-0.14
riba
-0.13
POSITIVE LOGITS
shar
0.14
Counter
0.14
savvy
0.14
warrants
0.14
icits
0.14
SCAN
0.14
tapes
0.14
accurate
0.13
curacy
0.13
typically
0.13
Activations Density 0.016%