INDEX
Explanations
words and phrases related to data processing and transfer, such as "transform," "send," "place," "transport," and "halt."
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1177
+0.16
0.5%
184
+0.12
0.4%
690
+0.11
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1531
+0.16
0.06
678
+0.12
0.06
690
+0.11
0.03
Negative Logits
كويكب
-0.69
Clik
-0.67
insuffisamment
-0.65
manuten
-0.64
minimalis
-0.64
XCTAssert
-0.64
alkoh
-0.63
akade
-0.62
MLLoader
-0.61
pels
-0.61
POSITIVE LOGITS
gaily
1.08
hairc
1.03
apprehen
0.99
unspeak
0.99
tolerably
0.97
plenti
0.97
disreg
0.97
impra
0.96
encomp
0.96
fucker
0.95
Activations Density 0.362%