INDEX

Explanations

code and command line syntax

The neuron detects numeric tokens that include decimal points (floating‐point numbers).

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

ной

0.73

IENT

0.71

 свого

0.70

aholic

0.69

 Serviço

0.69

рачи

0.68

 Allgemeinen

0.68

ASH

0.67

 хол

0.66

 தனது

0.66

POSITIVE LOGITS

بی

0.79

Như

0.78

冽

0.78

 posljed

0.76

 craindre

0.76

Мы

0.75

Č

0.74

ủi

0.73

śc

0.72

 úgy

0.71

Activations Density 0.000%