INDEX

Explanations

linking or connecting evidence

The neuron activates on legal-proceedings jargon—particularly terms naming or referring to elements of a court case (e.g. “alibi,” “identification,” “evidence,” “defense,” “case,” etc.).

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

 skriv

-1.24

 same

-1.18

 geene

-1.16

 marte

-1.15

pullover

-1.15

 tark

-1.14

 organisme

-1.11

 tena

-1.10

 dezelve

-1.09

 apenas

-1.09

POSITIVE LOGITS

"])){

1.23

()){

1.23

 Einfluß

1.17

'])){

1.16

StatusLabel

1.16

</strong>

1.15

 that

1.15

 chronologically

1.13

čít

1.12

 sartén

1.09

Activations Density 0.034%