INDEX

Explanations

The neuron responds to the double‐equals comparison operator (“==”) in code.

case statements

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Août

-1.11

 déput

-1.09

 artesanato

-1.07

batu

-1.07

 magnifiques

-1.03

 vendredi

-1.01

 mít

-0.98

 cabeça

-0.98

鶫

-0.97

仕上が

-0.96

POSITIVE LOGITS

or

1.27

妻子

0.97

Pozri

0.95

if

0.94

 насеља

0.92

❤

0.91

isNot

0.90

 then

0.88

 mehrm

0.87

 kült

0.86

Activations Density 0.001%