INDEX

Explanations

regex emails, channel, gallery, contacts

The neuron strongly activates on numeric tokens (e.g. amounts, percentages, and other number strings).

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

 nacimiento

0.74

ное

0.71

bodyA

0.70

 diverg

0.68

 conoce

0.68

 colg

0.68

थरूम

0.67

্টর

0.67

एक

0.66

 corta

0.65

POSITIVE LOGITS

 สิน

0.86

ยาว

0.78

CURRENT

0.74

 menacing

0.74

Kons

0.73

in

0.72

罨

0.72

zar

0.71

 impass

0.71

ราย

0.71

Activations Density 0.001%