INDEX

Explanations

names ending in er or man

The neuron detects capitalized name fragments or tokens that are part of person or organization names.

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

of

-1.62

-1.56

-1.42

-1.41

-1.30

but

-1.30

-1.29

),

-1.27

従来

-1.27

POSITIVE LOGITS

 estadounid

1.63

 parachoque

1.61

 cewek

1.58

ourcen

1.56

UnitTesting

1.51

ᚐ

1.51

 kesel

1.48

 kule

1.48

鲧

1.47

偁

1.47

Activations Density 0.146%