INDEX

Explanations

of the words or phrases

The neuron activates on documentation-style technical terms referring to HTTP/API data structures—words like “corpo,” “caracteres,” “argumento,” “request,” “response,” and “streaming.”

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

 exportaciones

-1.55

 кілька

-1.52

 acoged

-1.42

 eenige

-1.34

 localizar

-1.31

 negociaciones

-1.30

釙

-1.30

 contratar

-1.29

 sentada

-1.28

 cantit

-1.27

POSITIVE LOGITS

of

8.81

ของ

3.97

 của

3.67

 της

3.06

 ofthe

2.80

 του

2.55

 של

2.53

 della

2.48

 των

2.44

ofthe

2.19

Activations Density 0.016%