INDEX
Explanations
The neuron flags mentions of monetary contract figures (e.g., dollar amounts and “million” values).
New Auto-Interp
Negative Logits
ümüz
-0.06
trom
-0.06
app
-0.06
mají
-0.06
Cowboy
-0.06
python
-0.06
İki
-0.06
brother
-0.06
Sanat
-0.05
_thresh
-0.05
POSITIVE LOGITS
-purpose
0.07
_ONCE
0.07
PHONE
0.07
rary
0.07
Manufacturer
0.06
ляет
0.06
корпус
0.06
'nde
0.06
рю
0.06
variations
0.06
Activations Density 0.005%