INDEX

Explanations

role of

The neuron primarily activates on numeric literals (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

to

-2.70

 gekomen

-1.45

 dieną

-1.41

 }}$,

-1.37

了几

-1.34

 from

-1.33

-1.31

 Particular

-1.27

via

-1.27

="#"><

-1.27

POSITIVE LOGITS

メイン

1.45

.”

1.40

 berge

1.38

我已经

1.34

“

1.32

卨

1.32

去年

1.30

…」

1.29

fusions

1.29

 milliard

1.27

Activations Density 0.005%