INDEX

Explanations

simplifying processes and workflows

This neuron responds to marketing and promotional language—especially second-person calls to action like “make,” “your,” “save,” and similar ad-style phrasing.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 privacy

0.56

 integrity

0.52

 cava

0.52

 meliputi

0.51

 подразуме

0.50

DID

0.49

 sensitivity

0.48

 completeness

0.48

alk

0.47

 해당

0.47

POSITIVE LOGITS

 жизнь

1.14

 życie

0.92

过程

0.89

 வாழ்க்க

0.86

transition

0.86

Transition

0.85

 процесс

0.85

 زندگی

0.84

 життя

0.84

 transitions

0.82

Activations Density 0.338%