INDEX

Explanations

medical terms, brand names

This neuron detects technical or specialized terms in various domains like medicine, sports, programming, and academic contexts.

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

Tamanho

-1.41

$=$

-1.34

toEqual

-1.28

.}

-1.27

 into

-1.25

 vigente

-1.25

Escrito

-1.24

廸

-1.23

 sentada

-1.23

Продам

-1.22

POSITIVE LOGITS

 these

1.80

not

1.77

 this

1.73

 which

1.61

 that

1.57

 whose

1.56

how

1.55

all

1.45

the

1.45

 those

1.44

Activations Density 0.107%