INDEX

Explanations

connecting specific concepts

The neuron activates on numeric tokens, especially measurements or statistical values (decimal and integer numbers).

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

-1.55

['

-1.52

">

-1.43

]-'

-1.43

('

-1.41

isierte

-1.32

٣

-1.28

过去了

-1.25

 cathédrale

-1.24

POSITIVE LOGITS

 such

1.46

its

1.42

of

1.41

 this

1.35

колай

1.29

֒

1.27

دين

1.24

на

1.23

بوب

1.23

père

1.23

Activations Density 0.246%