INDEX

Explanations

to my or the

This neuron identifies common function words—especially prepositions like “to” and frequent pronouns/possessives such as “my,” “our,” and articles like “the.”

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

 Then

-1.52

 would

-1.51

 this

-1.42

 Since

-1.42

 then

-1.38

ueuse

-1.37

as

-1.35

 during

-1.34

likace

-1.34

 could

-1.32

POSITIVE LOGITS

在那里

1.64

どうして

1.50

 desperdic

1.47

 ..."

1.45

 komplet

1.43

 henvisninger

1.42

ynchronously

1.42

possibly

1.42

ständig

1.41

ynchronous

1.39

Activations Density 0.029%