INDEX

Explanations

stipulated in legal texts

The neuron responds to numeric references and legal/regulatory citations (e.g. years, section numbers, CFR references, code citations).

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

OTROS

-1.16

rigada

-1.08

 wasn

-1.04

UARIO

-1.02

taje

-1.02

堯

-1.01

 vielleicht

-1.01

bestos

-1.00

 infractions

-1.00

 nač

-1.00

POSITIVE LOGITS

law

1.26

 guidelines

1.23

МЕН

1.16

in

1.16

 prawda

1.15

 their

1.13

the

1.11

 legislation

1.09

행

1.04

 förra

1.03

Activations Density 0.032%