INDEX

Explanations

consequent increase or lower bound

The neuron activates on numeric values (integers or decimals) in technical or scientific 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

his

-1.40

 ilyen

-1.35

はこの

-1.33

mands

-1.33

</b>

-1.33

 suoi

-1.32

 ônibus

-1.31

 helpt

-1.30

 livro

-1.30

 veículo

-1.26

POSITIVE LOGITS

or

1.59

 вместо

1.36

 кількість

1.30

cillary

1.27

最后由

1.25

my

1.22

if

1.15

 zamiast

1.12

めで

1.10

кото

1.10

Activations Density 0.116%