INDEX

Explanations

code delimiters and terminators

The neuron fires on numeric literal tokens (i.e. numbers) in the code.

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

 только

-1.44

 biblia

-1.41

不明白

-1.30

 that

-1.30

を使用した

-1.27

 kupa

-1.27

).}

-1.26

 temen

-1.25

 profeta

-1.21

 didik

-1.20

POSITIVE LOGITS

to

1.56

 usual

1.34

四十

1.28

 поча

1.27

in

1.27

 begin

1.27

 hoping

1.26

ڧ

1.26

ways

1.25

 began

1.23

Activations Density 0.009%