INDEX

Explanations

street, titles, names, locations

The neuron fires on tokens that occur immediately adjacent to or as part of floating-point (decimal) numerical values.

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

and

-2.14

ņas

-1.20

 from

-1.20

 conçu

-1.16

もし

-1.15

 сегодня

-1.13

甚至

-1.13

 duquel

-1.13

-1.10

 има

-1.07

POSITIVE LOGITS

to

1.77

 they

1.66

 akci

1.45

 zeggen

1.45

:=\

1.45

նակ

1.40

那你

1.38

clairage

1.32

 menik

1.30

or

1.28

Activations Density 0.182%