INDEX

Explanations

English

The neuron detects property listing headlines and key numeric real-estate specs (e.g. “2 bedroom apartment,” location, price).

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

觧

-1.83

だけではなく

-1.63

だけでなく

-1.60

 bayern

-1.55

੦

-1.55

甭

-1.55

歟

-1.54

 kompeti

-1.54

も見

-1.52

 rezept

-1.51

POSITIVE LOGITS

1.77

 like

1.70

has

1.65

in

1.57

but

1.53

is

1.48

 mostly

1.45

 mainly

1.39

靱

1.39

Activations Density 0.073%