INDEX

Explanations

listen on port

The neuron detects words and identifiers related to a service listening or serving on a port (e.g. “listen,” “escutar,” “Server,” “listener,” “static”).

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

 such

-1.40

if

-1.29

 when

-1.20

 about

-1.17

 に関する

-1.15

 that

-1.13

ご了承

-1.06

ліза

-1.06

ποίη

-1.03

ご了承ください

-1.02

POSITIVE LOGITS

诡异

1.34

1.23

1.19

 nuclé

1.16

vó

1.09

 monstre

1.09

礴

1.08

那人

1.07

Anak

1.03

ioneers

1.02

Activations Density 0.001%