INDEX

Explanations

extending, expanding, widening

The neuron detects terms related to making something larger or resizing, such as “expand,” “extend,” “shrink,” or “resizing.”

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 навіть

-1.41

 éstas

-1.38

Generally

-1.30

 ПОД

-1.27

 действует

-1.26

era

-1.22

That

-1.21

 ДЕ

-1.18

Even

-1.17

邊

-1.16

POSITIVE LOGITS

the

2.05

fotografie

1.49

its

1.48

his

1.45

২

1.38

卺

1.32

 preocupado

1.31

endimiento

1.27

 قبل

1.26

Teknik

1.23

Activations Density 0.092%