INDEX

Explanations

Here with

The neuron fires strongly on numerals and list‐style headings (e.g. digits marking sections or bullet points).

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

 genannten

-1.37

 cũng

-1.34

 также

-1.31

 także

-1.30

 også

-1.29

 same

-1.28

also

-1.27

 also

-1.25

 Boletín

-1.23

 グラフ

-1.23

POSITIVE LOGITS

unikasi

1.34

 suena

1.27

 dAtA

1.27

 hvilke

1.23

new

1.23

 forskj

1.23

 have

1.22

 sekunder

1.21

 enquanto

1.20

 våre

1.20

Activations Density 0.345%