INDEX

Explanations

legal claims and actions

The neuron fires on formal legal procedure terms—especially words naming or describing court actions (e.g. warrants, arrests, charges) in judicial opinions.

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

 chrétienne

-1.53

 çünkü

-1.45

larımız

-1.39

 yalnızca

-1.36

 teórico

-1.36

 extremadamente

-1.36

心底

-1.30

 sobretudo

-1.30

 блюдо

-1.29

formazioni

-1.27

POSITIVE LOGITS

 лечения

1.27

 reale

1.26

濛

1.21

was

1.19

 complained

1.18

 comunit

1.17

하였다

1.16

ả

1.16

hagic

1.15

 Kochen

1.15

Activations Density 0.073%