INDEX
Explanations
The neuron is primarily responding to sentence-ending punctuation (periods) in the text.
New Auto-Interp
Negative Logits
ฟร
-0.07
replicate
-0.07
�
-0.07
.Please
-0.07
tolerance
-0.06
Flood
-0.06
-angular
-0.06
_BO
-0.06
RTOS
-0.06
_term
-0.06
POSITIVE LOGITS
quân
0.06
’den
0.06
становится
0.06
.controls
0.06
ücü
0.06
conson
0.06
Дон
0.06
_Renderer
0.06
ซ
0.05
.modal
0.05
Activations Density 0.008%