INDEX

Explanations

function call parameters

The neuron fires on numeric literal tokens (integer constants) in code.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

if

-1.79

for

-1.65

 after

-1.48

 other

-1.45

 when

-1.42

<bos>

-1.41

by

-1.39

 with

-1.34

 komplette

-1.34

to

-1.31

POSITIVE LOGITS

 quels

1.41

 اطلع

1.34

 increíbles

1.33

 ſont

1.26

 actuels

1.25

 impresionantes

1.25

 SCHOLASTIC

1.25

え

1.22

 quelles

1.21

 nationaux

1.21

Activations Density 0.002%