INDEX

Explanations

of [nouns]

The neuron strongly responds to marketing‐style, promotional quality descriptors and superlative adjectives (e.g. “ideal,” “superior,” “expertise,” “attained”) used to hype products or services.

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

ごちそう

-1.41

 suéter

-1.36

 encantador

-1.36

໊

-1.30

 distribuidor

-1.29

 causado

-1.29

使った

-1.29

 dzied

-1.29

佷

-1.27

思います

-1.27

POSITIVE LOGITS

1.47

</em>

1.34

in

1.31

 عزیزم

1.30

on

1.27

 there

1.22

 ristorante

1.22

 paradiso

1.20

ibi

1.20

бель

1.20

Activations Density 0.043%