INDEX

Explanations

property descriptions

The neuron is specialized to detect mentions of the word “property,” as used in real‐estate listings.

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

which

-1.85

and

-1.60

or

-1.58

\&

-1.55

-1.52

where

-1.50

-1.45

-1.40

—

-1.38

."

-1.38

POSITIVE LOGITS

,,,

2.03

 bedingt

2.00

냑

1.89

᠁

1.87

you

1.85

 hervorragend

1.83

 frág

1.81

нимание

1.80

 individuellen

1.79

to

1.74

Activations Density 0.014%