INDEX

Explanations

divided into parts

The neuron detects language describing the splitting or dividing of something into parts or sections (e.g., “split,” “divided,” “sections”).

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

 when

-1.69

思います

-1.43

 despite

-1.38

 before

-1.25

 perhaps

-1.22

see

-1.20

 without

-1.20

 support

-1.14

one

-1.12

 during

-1.12

POSITIVE LOGITS

 into

1.48

 olika

1.34

 différents

1.31

 poszczegól

1.27

ほら

1.23

 Publicado

1.23

űr

1.16

 ハーフ

1.16

();*/

1.16

 fendi

1.15

Activations Density 0.038%