INDEX

Explanations

`think`, `look`, `when`, `what`

The neuron strikes on prominent clause-level tokens—especially personal pronouns (I, you, he, etc.) and key main verbs or question words that signal the start of a proposition.

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

 will

-1.84

-1.54

-1.50

-1.45

cloudf

-1.34

and

-1.29

-1.27

 όλα

-1.25

 walaupun

-1.23

—

-1.23

POSITIVE LOGITS

it

1.47

ー

1.38

⎯

1.37

вление

1.34

ház

1.34

 griega

1.34

we

1.31

when

1.30

 tragedia

1.30

 extranjera

1.28

Activations Density 0.019%