INDEX

Explanations

older properties and houses

The neuron broadly lights up on substantive content words—nouns, verbs, adjectives—rather than on common function words.

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

who

-1.44

hores

-1.38

 your

-1.37

}(\

-1.34

 оружи

-1.33

ofing

-1.33

Dữ

-1.31

zensionen

-1.31

tủ

-1.30

 duración

-1.29

POSITIVE LOGITS

1.32

 النبات

1.28

tries

1.25

 elevador

1.23

 corredor

1.23

 roślin

1.20

 satélite

1.20

ʽ

1.16

хии

1.13

of

1.12

Activations Density 0.039%