INDEX

Explanations

designed

The neuron chiefly detects numeric expressions—tokens representing numbers, measurements, counts, or other digit‐based data.

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

够了

-1.53

-1.47

ранее

-1.47

 Dijo

-1.34

 barrage

-1.28

明確

-1.27

 gaily

-1.26

-¿

-1.23

 wodurch

-1.23

 mutter

-1.22

POSITIVE LOGITS

1.77

 that

1.42

 design

1.34

設計

1.34

 også

1.27

 itself

1.25

 sehat

1.25

 rests

1.23

its

1.22

sensory

1.22

Activations Density 0.055%