INDEX

Explanations

describes state or characteristic

The neuron fires on multiword verb–preposition patterns (e.g. “conforms to,” “transformed into,” “updated when,” “pouring onto”) that describe how something is related or changed.

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

in

-1.72

if

-1.70

are

-1.70

 although

-1.48

but

-1.43

had

-1.40

and

-1.35

 Mün

-1.30

by

-1.28

 massive

-1.23

POSITIVE LOGITS

 tiende

1.82

 stattfinden

1.62

 vantagens

1.57

 frutto

1.55

 giocatore

1.48

>\<

1.48

 souvent

1.48

 malheure

1.46

 parfois

1.45

怎样的

1.45

Activations Density 0.460%