INDEX
Explanations
expressions indicating physical distance or separation
far, distant, or apart
New Auto-Interp
Negative Logits
BagConstraints
-0.57
makeStyles
-0.49
noDo
-0.48
Presidencia
-0.47
-0.46
ActionCreators
-0.46
hyrchwyd
-0.44
realizarse
-0.44
bluzka
-0.44
désol
-0.43
POSITIVE LOGITS
nearby
0.65
Nearby
0.64
distant
0.59
nearby
0.59
near
0.56
distant
0.55
Near
0.54
far
0.54
Nearby
0.54
featureID
0.53
Activations Density 0.004%