INDEX

Explanations

numbers and measurements

This neuron activates on numeric tokens (digits or numerical measurement values).

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

 this

-2.22

 while

-2.17

 which

-2.05

 even

-1.93

if

-1.92

 only

-1.89

 before

-1.89

on

-1.88

of

-1.88

 that

-1.86

POSITIVE LOGITS

 hadde

1.72

踅

1.63

 cuatro

1.62

 feine

1.60

those

1.56

𝐫

1.52

,,

1.52

 sette

1.49

owns

1.48

dix

1.48

Activations Density 0.002%