INDEX
Explanations
requests for subscription or login
commas in the text, indicating pauses or separations in phrases
New Auto-Interp
Negative Logits
================================================================
-0.55
åħī
-0.54
é¾įå¥ij士
-0.54
fter
-0.53
ipel
-0.53
naire
-0.53
Footnote
-0.51
âĢİ
-0.51
ãĥį
-0.50
ccording
-0.50
POSITIVE LOGITS
try
0.76
submit
0.70
subscribe
0.64
please
0.63
Login
0.63
login
0.62
click
0.59
ancel
0.57
Try
0.57
submit
0.56
Activations Density 0.013%