INDEX

Explanations

This neuron seems to activate on documentation comments, exponents in scientific notation, and table formatting.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 also

-0.77

IsContent

-0.71

also

-0.69

 też

-0.62

 также

-0.60

 também

-0.59

 idem

-0.59

IMPORTED

-0.58

 también

-0.57

 الدولى

-0.56

POSITIVE LOGITS

RegressionTest

0.93

:✨

0.90

SourceChecksum

0.81

fjspx

0.77

 autorytatywna

0.71

antwoorde

0.70

зулта

0.68

 للاسماء

0.66

ThemeOverlay

0.65

 betweenstory

0.64

Activations Density 0.001%