INDEX
Explanations
references to the concept of "sprouting" or growth, particularly in a metaphorical or abstract context
New Auto-Interp
Negative Logits
Ih
-0.17
soap
-0.15
ê²
-0.15
XL
-0.15
uring
-0.15
spit
-0.14
rust
-0.14
Soap
-0.14
XL
-0.14
iever
-0.13
POSITIVE LOGITS
spr
0.28
uce
0.27
Spr
0.26
spr
0.24
Spr
0.23
ague
0.21
cial
0.21
outs
0.21
inkle
0.21
ightly
0.20
Activations Density 0.010%