INDEX

Explanations

code output

The neuron strongly activates on numeric literal tokens (numbers and numeric constants) in code or text.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 时尚

1.51

 ベージュ

1.39

 那麼

1.37

 コットン

1.31

綿

1.31

 ベルト

1.30

 पंतप्रधान

1.30

 ໜອງ

1.30

 通勤

1.30

zechoslovakia

1.27

POSITIVE LOGITS

if

2.72

print

1.91

//

1.83

if

1.82

for

1.82

return

1.77

try

1.55

console

1.55

NSLog

1.51

String

1.48

Activations Density 0.001%