INDEX

Explanations

family

The neuron is triggered by mentions of “family” (often in contexts like “family and friends”).

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

}$,

-1.13

to

-1.04

 verur

-1.02

}$.

-1.00

 веса

-0.99

 уже

-0.98

)}$,

-0.97

in

-0.96

}$;

-0.95

𝜙

-0.94

POSITIVE LOGITS

 itali

1.13

ributed

1.11

 lentement

1.05

bew

1.04

 lanterna

1.02

ıyorum

1.02

 autori

1.01

pueden

1.01

肯定会

0.99

 zamiast

0.99

Activations Density 0.001%