INDEX

Explanations

limit notation

The neuron fires on tokens that are numeric literals—i.e. multi‐digit numbers appearing in equations or limit expressions.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

bastian

-0.76

仍

-0.76

JAMIN

-0.74

deve

-0.73

usschuss

-0.72

atán

-0.72

 verbs

-0.72

 العاصمة

-0.71

givers

-0.71

 シューズ

-0.70

POSITIVE LOGITS

⁡

0.91

┬

0.84

 MAKE

0.79

 tienda

0.78

xm

0.76

CHURCH

0.75

nowa

0.75

 réflex

0.75

 expériment

0.74

 уход

0.72

Activations Density 0.010%