INDEX

Explanations

placement and location

The neuron fires on past‐tense verbs that describe physically placing, inserting, positioning, or mounting objects (e.g. inserted, placed, mounted, removed, stored) in a procedural or technical context.

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

-1.38

 χωρίς

-1.30

fsen

-1.29

to

-1.28

 längere

-1.28

-1.27

нала

-1.17

↵

-1.17

 витра

-1.17

oded

-1.16

POSITIVE LOGITS

 then

1.64

妧

1.45

unser

1.38

مرا

1.35

 auffi

1.30

῞

1.23

our

1.23

ڤ

1.23

 destinos

1.21

 whoſe

1.20

Activations Density 0.390%