INDEX

Explanations

numerous or many

The neuron strongly activates on the formal, specification-style language and structure typical of patent or technical disclosure text.

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

 autre

-1.47

哧

-1.41

𝟵

-1.40

 joue

-1.35

ophe

-1.33

 sosteniendo

-1.32

 luchando

-1.29

毙

-1.28

 aynı

-1.27

AVAILABLE

-1.20

POSITIVE LOGITS

Моя

1.27

There

1.24

it

1.23

ญี่ป

1.22

Numerous

1.22

Sinopsis

1.21

Razer

1.21

 มัน

1.20

GIUM

1.18

許多

1.17

Activations Density 0.413%