INDEX
Explanations
The neuron fires on occurrences of the stem “ischemi-,” i.e. words referring to ischemia or ischemic.
New Auto-Interp
Negative Logits
deposits
-0.07
(T
-0.07
Walters
-0.07
Tat
-0.06
Bon
-0.06
Bart
-0.06
FU
-0.06
pets
-0.06
tal
-0.06
ro
-0.06
POSITIVE LOGITS
ischem
0.17
іка
0.07
IRM
0.07
Shim
0.07
reeze
0.07
شرقی
0.07
сих
0.07
lstm
0.07
Isabel
0.07
Erica
0.07
Activations Density 0.003%