INDEX

Explanations

so that, ensure

The neuron activates strongly on purpose‐clauses and enabling phrases—especially “so that,” “allow … to,” and modals like “can”—marking text that expresses why or how something is done.

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

-1.43

ꈍ

-1.23

by

-1.22

on

-1.21

codiles

-1.20

ceptre

-1.17

 aneh

-1.16

ouchers

-1.15

 mereka

-1.14

 tersenyum

-1.14

POSITIVE LOGITS

鰱

1.40

早速

1.27

久しぶりの

1.26

さっそく

1.24

 then

1.23

bemos

1.20

久しぶりに

1.20

&&(

1.18

ਈ

1.18

 gettyimages

1.14

Activations Density 0.056%