INDEX
Explanations
property
This neuron activates on terms referring to a company’s service commitments and quality assurances (e.g., “service,” “security,” “sustainability,” “maintenance”).
New Auto-Interp
Negative Logits
decoder
-0.07
кие
-0.06
ullah
-0.06
ка
-0.06
tròn
-0.06
.minute
-0.06
picker
-0.06
_CHARS
-0.06
teg
-0.06
plc
-0.06
POSITIVE LOGITS
rowsing
0.07
介
0.07
_tf
0.07
变
0.06
attles
0.06
'''↵↵
0.06
'<
0.06
Olympics
0.06
об
0.06
ทำงาน
0.06
Activations Density 0.094%