INDEX

Explanations

relates to / pertains to / defined by

The neuron detects verbs that express a relation or reference—especially forms like “pertain(s) to,” “relate(s) to,” “appear(s) in,” or “apply(ies) to.”

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

 DELHI

-1.27

♔

-1.26

 directiva

-1.25

 Bildung

-1.24

volvement

-1.23

čnost

-1.23

 fantastique

-1.23

⢓

-1.22

 تجهیزات

-1.22

 QtGui

-1.22

POSITIVE LOGITS

 this

1.75

on

1.64

 both

1.58

 these

1.55

 those

1.37

johnson

1.35

 zarówno

1.35

our

1.34

 your

1.34

1.30

Activations Density 0.064%