INDEX

Explanations

"**" or "("

The neuron is detecting interrogative “What is…?” questions that ask for specific attributes (e.g. profession, birth place, height, etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 massac

0.58

 أيضاً

0.53

0.48

 notific

0.48

żenia

0.47

 estés

0.47

ған

0.47

.';

0.46

!”.

0.46

POSITIVE LOGITS

🤔

0.56

⏎

0.55

해요

0.54

()=>{

0.53

ето

0.52

0.51

いない

0.51

어요

0.50

ㅎㅎ

0.50

haust

0.49

Activations Density 0.002%