INDEX
Explanations
words related to measurements in the context of temperature
New Auto-Interp
Negative Logits
ray
-0.16
ester
-0.15
atis
-0.15
chu
-0.15
iate
-0.15
elle
-0.15
arios
-0.15
èµı
-0.14
imps
-0.14
igo
-0.14
POSITIVE LOGITS
ONSE
0.16
artz
0.16
ipur
0.15
STYPE
0.15
ngang
0.14
quette
0.14
ATUS
0.14
freelance
0.14
alim
0.14
imizer
0.14
Activations Density 0.010%