INDEX
Explanations
the word "desperate" and its variations, indicating a focus on themes of urgency and need
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
156
+0.20
1.1%
381
+0.10
0.6%
327
+0.10
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
226
+0.20
0.02
156
+0.10
0.02
432
+0.10
0.02
Negative Logits
Said
-1.65
Month
-1.58
riel
-1.51
Foundation
-1.46
columnist
-1.45
EEK
-1.45
diversity
-1.43
roma
-1.42
ierno
-1.39
ISH
-1.38
POSITIVE LOGITS
Īĺ
4.14
ĥ
3.73
¼
3.58
ĻĤ
3.57
Ħ
3.47
Ĥ¬
3.44
ħ
3.31
ij
3.27
°
3.20
¸
3.20
Activations Density 0.057%