INDEX
Explanations
attends to updates marked by asterisks from earlier unchanged code lines
New Auto-Interp
Head Attr Weights
0:0.07
1:0.10
2:0.56
3:0.08
4:0.04
5:0.02
6:0.03
7:0.06
Negative Logits
generalization
-0.31
numerusform
-0.27
tenti
-0.26
cap
-0.26
materia
-0.26
cap
-0.25
chart
-0.25
:
-0.25
estr
-0.25
Donahue
-0.25
POSITIVE LOGITS
}}"></
0.52
tartalomajánló
0.45
')))
0.44
]<<"
0.43
')));
0.43
"]));
0.43
']));
0.42
'}>
0.42
"]];
0.42
}>;
0.42
Activations Density 0.126%