INDEX

Explanations

question followed by numbers or we

The neuron is triggered by numeric tokens (e.g. years, volume/page numbers, docket or citation numbers) in the text.

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

and

-1.80

çük

-1.35

 aluminio

-1.33

ʁ

-1.33

nó

-1.32

ści

-1.30

 своё

-1.29

செ

-1.27

生素

-1.27

felves

-1.26

POSITIVE LOGITS

 будут

1.65

 כמה

1.49

 those

1.34

いただきたい

1.30

how

1.30

 their

1.30

 some

1.28

are

1.27

 debe

1.26

 such

1.23

Activations Density 0.010%