INDEX
Explanations
Keywords related to fish fins or physical attributes like weight
references to fins or fin-like structures
New Auto-Interp
Negative Logits
Dialogue
-0.76
erity
-0.74
Adapt
-0.72
Pope
-0.70
Alert
-0.70
enza
-0.69
rika
-0.69
ERA
-0.68
rouch
-0.67
rians
-0.67
POSITIVE LOGITS
nets
0.88
emouth
0.82
eness
0.79
uit
0.77
hee
0.77
oths
0.77
gey
0.75
shed
0.74
essed
0.73
oise
0.73
Activations Density 0.045%