INDEX

Explanations

coefficient matrix, you, not

The neuron strongly fires on numeric or measurement tokens—especially digits, decimal values, and other precise quantitative markers.

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

 oferecer

0.92

 poderá

0.91

銇

0.90

 compétences

0.88

 soluções

0.88

 అనేక

0.87

னுக்கு

0.86

ColorEffects

0.86

 сроки

0.86

ہنی

0.85

POSITIVE LOGITS

id

0.90

sd

0.82

or

0.82

0.79

0.78

db

0.76

grp

0.75

ers

0.72

Activations Density 0.000%