INDEX

Explanations

taking photos

The neuron activates on words referring to photographing—e.g. “photo,” “picture(s),” “take,” “snap,” “photos.”

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

作って

-1.53

-1.28

 senin

-1.22

!」

-1.21

 المعيار

-1.21

 indignant

-1.20

 ambivalent

-1.19

 creada

-1.16

 portatil

-1.16

changers

-1.14

POSITIVE LOGITS

举例

1.39

 rôles

1.37

 rimanere

1.34

шёл

1.34

 iniziare

1.32

 because

1.29

摠

1.28

 bonitos

1.27

𝓶

1.27

Activations Density 0.040%