INDEX
Explanations
references to "lens" or "prism" as metaphors for perspective
New Auto-Interp
Head Attr Weights
0:0.02
1:0.01
2:0.08
3:0.06
4:0.12
5:0.02
6:0.05
7:0.36
8:0.04
9:0.04
10:0.07
11:0.08
Negative Logits
guard
-1.53
keys
-1.52
guards
-1.51
keyboards
-1.46
itars
-1.43
barric
-1.42
joined
-1.42
sacks
-1.41
typew
-1.39
pian
-1.38
POSITIVE LOGITS
Venture
1.54
abal
1.48
inka
1.44
��
1.41
Inquiry
1.41
Higher
1.40
gradient
1.37
Charity
1.35
Profit
1.33
Vision
1.31
Activations Density 0.001%