INDEX
Explanations
This neuron detects occurrences of the “Mozilla Public License” text in source‐code headers.
New Auto-Interp
Negative Logits
non
-0.07
aniel
-0.06
ir
-0.06
Mansion
-0.06
Ro
-0.06
NON
-0.06
IF
-0.06
Gol
-0.06
:]↵
-0.06
�
-0.06
POSITIVE LOGITS
емого
0.07
креп
0.07
nuts
0.07
темп
0.07
unclear
0.07
billing
0.07
áf
0.07
NUnit
0.06
кон
0.06
yeah
0.06
Activations Density 0.001%