INDEX

Explanations

not enough or more

The neuron flags negation or limiting constructions—words and phrases that express “not,” “not enough,” “not all,” etc.

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

 вижу

-1.50

not

-1.47

 říká

-1.45

now

-1.42

with

-1.36

 приходит

-1.35

 видел

-1.34

Răsp

-1.33

 costruire

-1.31

 giving

-1.30

POSITIVE LOGITS

of

1.67

que

1.41

 ifølge

1.38

 tegens

1.34

 horrid

1.28

也不会

1.26

 stup

1.25

 SIGNIFIC

1.25

 weiteren

1.24

 hein

1.24

Activations Density 0.010%