INDEX

Explanations

biographies and descriptions

The neuron detects document section headings (e.g. “Life,” “Biography,” “Description,” “Overview”).

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

-1.58

 jedoch

-1.32

 Alquiler

-1.30

-1.28

of

-1.27

了不少

-1.26

 tercih

-1.26

eight

-1.24

『

-1.23

POSITIVE LOGITS

gemeen

1.52

鄖

1.41

hadapi

1.39

össä

1.39

维基

1.34

 ikkje

1.33

 allerede

1.31

𖡼

1.30

bodyParser

1.30

 alene

1.30

Activations Density 0.078%