INDEX

Explanations

must-have or essential items

The neuron picks out words and phrases that express necessity or strong recommendation (e.g. “must,” “must-have,” “essential”).

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

躄

-1.62

蜴

-1.60

htb

-1.52

 únicos

-1.50

 seleccionamos

-1.49

𐤔

-1.48

 Diallo

-1.45

 Перейти

-1.43

 všechny

-1.41

ųjų

-1.40

POSITIVE LOGITS

for

1.54

1.52

in

1.44

1.42

1.41

ции

1.40

也不能

1.39

so

1.38

It

1.37

 azzurro

1.34

Activations Density 0.025%