INDEX

Explanations

basics explained with numbers

The neuron strongly activates on numerical tokens (multi‐digit numbers) regardless of context.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

CONSE

-0.79

隹

-0.77

ᕐ

-0.74

SEVEN

-0.73

giers

-0.73

šnje

-0.73

 cuarent

-0.73

 تاپ

-0.72

 torrent

-0.72

LOOKING

-0.72

POSITIVE LOGITS

 basics

1.75

 Basics

1.50

basics

1.42

 básico

1.40

 lesson

1.33

 fundamentals

1.28

 básica

1.25

Basics

1.22

 Fundamentals

1.20

 explained

1.09

Activations Density 0.044%