INDEX

Explanations

population and age demographics

This neuron activates on identifier-like tags and labels (e.g. XML “name” attributes, “preview,” “documentation,” “verify,” etc.), i.e. tokens that serve as resource or section identifiers in code or markup.

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

 creates

-1.20

 creating

-1.14

 first

-1.13

 dozens

-1.07

 Vierge

-1.06

 nestled

-1.03

 combines

-1.02

 hundreds

-1.02

 provide

-1.00

 promoting

-0.99

POSITIVE LOGITS

 aged

1.58

of

1.38

age

1.34

 ages

1.30

oft

1.17

 older

1.16

 opulent

1.12

 ageing

1.09

≥

1.05

both

1.04

Activations Density 0.001%