INDEX

Explanations

across the world

The neuron responds strongly to spatial prepositions (like “in,” “over”) marking location or distribution.

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

 then

-1.48

 fantastique

-1.30

 than

-1.30

なのですが

-1.29

出して

-1.25

 joyería

-1.24

 skar

-1.23

りますが

-1.23

 soldats

-1.22

 joyas

-1.22

POSITIVE LOGITS

不同

1.48

1.47

 olika

1.40

 different

1.39

The

1.31

 cilvē

1.25

Our

1.25

 unterschied

1.22

<blockquote>

1.20

Activations Density 0.113%