INDEX

Explanations

mentions of age

This neuron detects numeric age references (numbers associated with “old,” “age,” or similar life‐stage terms).

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

</tfoot>

-1.48

rolex

-1.41

GraphicsUnit

-1.37

bkz

-1.35

šení

-1.30

bricas

-1.27

lumber

-1.27

ܲ

-1.26

łaś

-1.25

 semblait

-1.24

POSITIVE LOGITS

1.66

1.46

 With

1.45

1.40

1.38

how

1.38

in

1.37

1.32

與

1.30

𓐍

1.30

Activations Density 0.028%