INDEX
Explanations
phrases related to specific products or items followed by a number indicating a level of satisfaction or recommendation
New Auto-Interp
Negative Logits
WARD
-0.77
nor
-0.72
mit
-0.67
GOODMAN
-0.64
CEPT
-0.64
jon
-0.63
signature
-0.61
ne
-0.61
adoptive
-0.61
Nazi
-0.61
POSITIVE LOGITS
ines
1.11
inations
1.00
ined
0.90
ine
0.88
ining
0.88
Ń·
0.87
ibrary
0.82
ago
0.80
agic
0.79
ython
0.78
Activations Density 1.031%