INDEX
Explanations
references to religious themes and figures
New Auto-Interp
Negative Logits
uhn
-0.15
reten
-0.14
DEV
-0.14
eturn
-0.14
ullo
-0.14
529
-0.14
429
-0.14
orgot
-0.13
argent
-0.13
Vader
-0.13
POSITIVE LOGITS
proph
0.43
prophet
0.38
prophecy
0.35
prop
0.32
prophets
0.31
Prophet
0.30
prediction
0.29
predictions
0.29
predicting
0.28
predictive
0.28
Activations Density 0.221%