INDEX
Explanations
statements and declarations
New Auto-Interp
Negative Logits
funktioniert
0.93
quirks
0.90
hipster
0.86
wirklich
0.86
paranoia
0.82
bukan
0.81
obsess
0.81
ingestion
0.80
isn
0.79
magically
0.79
POSITIVE LOGITS
congratulated
1.47
commended
1.39
отметил
1.38
urged
1.36
manifestó
1.25
expresó
1.25
remarked
1.25
thanked
1.24
congrat
1.24
reiterated
1.23
Activations Density 0.032%