INDEX
Explanations
words related to heat and the act of swimming
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
378
+0.09
0.3%
595
+0.08
0.2%
964
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
342
+0.09
0.03
595
+0.08
0.03
1441
+0.08
0.03
Negative Logits
dichi
-0.76
Ottobre
-0.69
?</
-0.66
fumo
-0.65
motiva
-0.59
Settembre
-0.58
Novembre
-0.58
migli
-0.56
Dimen
-0.55
Marzo
-0.55
POSITIVE LOGITS
heat
0.79
summer
0.70
heat
0.68
temperatures
0.67
temperature
0.63
scorching
0.60
hotter
0.60
hot
0.60
Heat
0.59
summer
0.58
Activations Density 0.155%