INDEX

Explanations

numerical quantities or measures

The neuron fires on technical terms that specify quantitative measures or dimensions (e.g. “dimensional,” “dimensions,” “len,” “vector,” “manifold”).

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

聍

-1.58

Wyd

-1.37

('');

-1.34

yczne

-1.33

ᨕ

-1.32

 sier

-1.31

 metales

-1.30

諳

-1.27

 действу

-1.26

 cuchara

-1.26

POSITIVE LOGITS

 that

2.34

it

2.08

to

1.92

gäng

1.34

 each

1.33

hängt

1.30

 otras

1.30

 След

1.27

tisgarh

1.23

ที่จะ

1.22

Activations Density 0.039%