INDEX
Explanations
the word "concept" within various contexts
references to the term "concept."
New Auto-Interp
Negative Logits
ificant
-0.68
attering
-0.67
Silence
-0.66
azar
-0.65
ndum
-0.63
istle
-0.63
minster
-0.61
volcanic
-0.61
eru
-0.61
rivers
-0.60
POSITIVE LOGITS
ually
1.51
ual
1.01
concept
0.95
uality
0.86
ional
0.83
SHIP
0.82
urally
0.81
Concept
0.81
concept
0.79
ical
0.78
Activations Density 0.015%