INDEX

Explanations

The neuron strongly activates on numeric tokens—particularly floating‐point numbers or other detailed numeric values.

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 மற்றும்

0.61

bzw

0.61

0.54

・

0.54

 iaitu

0.53

-)

0.52

www

0.52

 yaitu

0.52

 های

0.51

 Zeitschr

0.51

POSITIVE LOGITS

dokter

0.58

 बोलेंगे

0.52

はもちろん

0.49

throttle

0.49

Kafka

0.47

signer

0.47

global

0.47

brightness

0.47

ರಿಕ

0.47

Aph

0.47

Activations Density 0.000%