INDEX
Explanations
references to a specific plant, particularly its medicinal and health-related properties
New Auto-Interp
Negative Logits
.gdx
-0.20
erland
-0.18
erd
-0.18
erness
-0.17
oad
-0.15
esc
-0.15
_gradients
-0.15
rawer
-0.15
emain
-0.14
oplayer
-0.14
POSITIVE LOGITS
ificial
0.25
ifacts
0.24
illery
0.18
ifice
0.17
otle
0.17
elon
0.17
icular
0.16
bitrary
0.15
isinin
0.15
istry
0.15
Activations Density 0.039%