INDEX

Explanations

i.e. introductions

The neuron activates on numeric tokens, especially measurements or enumerated values (e.g. decimal numbers, list indices, and counts).

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

 this

-1.77

it

-1.55

 will

-1.45

 этом

-1.39

 seems

-1.30

 might

-1.24

or

-1.23

 allows

-1.23

 этого

-1.21

 there

-1.20

POSITIVE LOGITS

notated

1.69

 kredit

1.66

糨

1.59



1.59

螃

1.53

 ส์

1.52

phro

1.43

 anglais

1.43

basically

1.43

 anniversaire

1.41

Activations Density 0.068%