INDEX

Explanations

UI elements and software terms

The neuron fires on technical noun‐phrases naming UI elements, code constructs, or document headings (e.g. “radio button,” “display item,” “menu,” “class,” “system variable,” “HTML Template,” etc.).

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

 unterstützt

-1.51

 berücksichtigen

-1.49

 breathtaking

-1.46

 vielfach

-1.45

 verschieden

-1.42

 bevorzug

-1.40

 jenen

-1.38

 umgangssprach

-1.38

 persön

-1.38

 mehrfach

-1.38

POSITIVE LOGITS

ઃ

1.35

and

1.30

in

1.27

1.17

not

1.16

 Very

1.16

 demikian

1.16

 Only

1.16

1.10

so

1.09

Activations Density 0.238%