INDEX

Explanations

bus company

The neuron detects tokens that are part of numeric information (e.g. numbers, measurements, route/time figures).

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

</h2>

-1.34

ığımız

-1.31

 this

-1.29

 orec

-1.28

pæ

-1.25

ẨM

-1.23

}}/>

-1.23

 hierdie

-1.23

形

-1.22

сыз

-1.20

POSITIVE LOGITS

颚

1.48

you

1.38

 roues

1.35

 campeonato

1.34

 réfrigérateur

1.33

 cyclisme

1.30

Kenapa

1.30

nhung

1.27

but

1.26

 najdete

1.26

Activations Density 0.007%