INDEX
Explanations
verbs following 't' or 'to'
New Auto-Interp
Negative Logits
myös
0.89
скорее
0.87
també
0.84
と思いました
0.83
prefers
0.81
también
0.79
também
0.79
也要
0.78
גם
0.78
כבר
0.76
POSITIVE LOGITS
necessarily
1.37
quite
1.11
bode
1.09
necessarily
1.03
necessariamente
1.03
appreciably
1.02
really
1.01
necesariamente
0.97
quite
0.96
adequately
0.93
Activations Density 0.121%