INDEX
Explanations
relatives
The neuron responds to kinship and close-relationship terms (e.g., brother, son, friend) often in possessive form.
New Auto-Interp
Negative Logits
-for
-0.07
庭
-0.07
_TREE
-0.07
ў
-0.06
clip
-0.06
HV
-0.06
_component
-0.06
靠
-0.06
_lt
-0.06
�
-0.06
POSITIVE LOGITS
Intern
0.07
,DB
0.06
Spokane
0.06
responseData
0.06
Dr
0.06
belirlen
0.05
Країна
0.05
dated
0.05
Gran
0.05
`{0.05
Activations Density 0.042%