INDEX
Explanations
words related to plans and schedules, and words that describe things that are new
New Auto-Interp
Negative Logits
h
-0.66
Samuels
-0.64
crapers
-0.62
visualiser
-0.61
ngel
-0.61
<eos>
-0.61
gemeinschaft
-0.59
Adamson
-0.59
sandwich
-0.59
(!)
-0.58
POSITIVE LOGITS
purpoſe
1.26
Jefus
1.23
Majefty
1.22
Efq
1.12
―――――
1.09
ſtate
1.09
Chriftian
1.05
myſelf
1.01
Houſe
1.01
Reſ
1.01
Activations Density 2.124%