INDEX

Explanations

content marketing, strategy, planning

This neuron detects numbered list markers and other outline-style elements (digits, hyphens, slashes) that indicate steps or enumerated headings.

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

 with

-2.14

but

-1.95

has

-1.86

 without

-1.77

 large

-1.72

had

-1.65

and

-1.59

 desagrad

-1.59

alami

-1.56

学部

-1.55

POSITIVE LOGITS

 your

2.69

Eſ

2.03

you

2.00

鼍

1.77

 görmek

1.77

 Inſ

1.75

 destinées

1.72

 ouver

1.71

 eorum

1.70

 verrou

1.69

Activations Density 0.580%