INDEX

Explanations

factors and attributes

The neuron responds to nouns denoting measurable or evaluative attributes (e.g. quality, length, intensity, price, perspective, atmosphere).

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

-1.57

+=

-1.42

 that

-1.39

 установка

-1.38

の中に

-1.37

da

-1.34

ités

-1.34

в

-1.33

 الجديد

-1.31

isait

-1.31

POSITIVE LOGITS

and

1.74

 noodzake

1.36

 innovación

1.34

eddah

1.30

 ancora

1.27

шения

1.27

JACK

1.27

 joyería

1.27

Sprache

1.27

 促销

1.27

Activations Density 0.217%