INDEX
Explanations
The neuron fires on numeric values and measurements (e.g. weights, carat counts, dimensions) in product descriptions.
New Auto-Interp
Negative Logits
次
-0.07
<tr
-0.07
bě
-0.06
�
-0.06
.tolist
-0.06
remarks
-0.06
card
-0.06
_phase
-0.06
cards
-0.06
پل
-0.06
POSITIVE LOGITS
웃
0.07
غراف
0.07
inconsist
0.07
κρι
0.06
Critical
0.06
طن
0.06
Expenses
0.06
еля
0.06
Firestore
0.06
христи
0.06
Activations Density 0.008%