EXPLANATIONS
say "fun" words
INFERENCE
IFRAME INTEGRATION
NEGATIVE LOGITS
loff-18.875
CAF-17.375
constants-17.125
姗-17.000
caff-16.250
_si-16.250
ystate-16.125
vertise-15.938
unread-15.875
reak-15.813
POSITIVE LOGITS
_fun12.938
_Functions12.375
HELL12.313
Ę12.000
لب12.000
луш11.938
_functions11.938
ENCES11.625
_FUNCTIONS11.563
ITICAL11.500
ACTIVATION TIMES 0.1887%
Activation histogram is not available.
Logits histogram is not available.
TOPACTIVATIONS
SUBSAMPLING80%
SUBSAMPLING60%
SUBSAMPLING40%
NONACTIVATING