INDEX

Explanations

disabled state

The neuron fires on tokens that specify an element’s enabled/disabled state—e.g. attribute names like “enabled” or “disabled.”

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

-1.91

↵

-1.56

 that

-1.28

 stratég

-1.25

,”

-1.23

 bụi

-1.22

 bisher

-1.16

kowej

-1.09

 your

-1.08

 geplant

-1.07

POSITIVE LOGITS

 Apalagi

1.34

Sekarang

1.34

 недостатки

1.33

="

1.32

 Novgorod

1.30

要是

1.30

־

1.29

 Beberapa

1.27

差異

1.27

Terima

1.25

Activations Density 0.022%