INDEX
Explanations
instances of the word 'being'
New Auto-Interp
Negative Logits
PsyNetMessage
-0.76
Dag
-0.69
inav
-0.65
robat
-0.65
Beginning
-0.61
ère
-0.59
eva
-0.59
sovere
-0.59
Surv
-0.59
Lag
-0.58
POSITIVE LOGITS
able
1.15
eaten
0.92
discussed
0.91
replaced
0.91
consumed
0.91
phased
0.88
touted
0.88
chased
0.87
hailed
0.87
considered
0.87
Activations Density 0.066%