INDEX
Explanations
words associated with food textures and qualities
New Auto-Interp
Negative Logits
igit
-0.18
urator
-0.16
gers
-0.16
IZED
-0.15
/YYYY
-0.15
tings
-0.14
ration
-0.14
çݰ
-0.14
Yard
-0.14
odcast
-0.14
POSITIVE LOGITS
y
1.17
iness
0.69
ier
0.67
iest
0.61
ily
0.61
yb
0.52
yn
0.48
yw
0.48
yg
0.46
IER
0.45
Activations Density 0.150%