INDEX

Explanations

The neuron selectively fires on numeric tokens (constants, exponents, coefficients) in mathematical formulas.

mathematical variables and operations

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

//};

-1.02

 Preschool

-0.88

눅

-0.82

 compétence

-0.81

 childrens

-0.81

 дописа

-0.81

缺點

-0.80

绉

-0.78

 отредакти

-0.78

 reactiv

-0.77

POSITIVE LOGITS

 then

1.09

棚

1.06

ersachsen

1.04

 также

1.01

۴

0.98

۱

0.94

 voltou

0.91

۳

0.90

 JOURNAL

0.88

CAST

0.86

Activations Density 0.028%