INDEX

Explanations

list items

This neuron detects tokens that are part of numerical, formatted, or metadata elements—such as dates, times, section numbers, measurements, and other structured alphanumeric identifiers.

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

 indulgent

0.95

 divertida

0.94

িয়েছেন

0.91

 adequada

0.89

還有

0.89

িয়েছে

0.87

 rxjs

0.86

arı

0.86

ной

0.85

 تكبرها

0.85

POSITIVE LOGITS

та

1.30

ल

1.28

ز

1.23

ка

1.05

м

1.04

РИ

1.03

م

1.02

ק

1.02

1.01

Activations Density 2.273%