EXPLANATIONS
[American] => [noun]
INFERENCE
IFRAME INTEGRATION
NEGATIVE LOGITS
痕-15.750
alin-14.938
nemonic-13.625
gli-13.125
äl-13.125
_ado-13.063
.debian-13.000
hee-12.563
_الرقم-12.438
unic-12.313
POSITIVE LOGITS
MaxY16.875
shots15.625
_NONNULL15.438
上去15.313
pios15.063
ersh15.000
SHOT14.625
缨14.500
炸14.500
_Crimes14.438
ACTIVATION TIMES 0.1793%
Activation histogram is not available.
Logits histogram is not available.
TOPACTIVATIONS
SUBSAMPLING80%
SUBSAMPLING60%
SUBSAMPLING40%
NONACTIVATING