INDEX

Explanations

wonder and questions

The neuron detects hedging or speculative language—modal and “wondering” phrases that signal uncertainty or conjecture.

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

 there

-1.63

you

-1.55

is

-1.41

 黄色

-1.36

the

-1.31

in

-1.27

 magasiner

-1.26

 concernés

-1.26

 более

-1.23

 román

-1.21

POSITIVE LOGITS

itinéraire

1.30

They

1.24

゙

1.23

𝐹

1.22

 нашел

1.21

？」

1.20

ți

1.20

YES

1.20

 wonder

1.19

dź

1.19

Activations Density 0.027%