INDEX

Explanations

module/class definitions

The neuron primarily detects floating‐point numeric literals (numbers containing a decimal point) in code.

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

↵

-1.80

and

-1.58

钾

-1.31

eez

-1.27

ポロ

-1.26

пъл

-1.26

 gjennom

-1.25

蛩

-1.24

 conteúdos

-1.24

ї

-1.23

POSITIVE LOGITS

its

1.63

one

1.52

are

1.38

 other

1.37

{}

1.32

the

1.27

லை

1.27

1.25

 ragazzo

1.23

no

1.23

Activations Density 0.002%