INDEX

Explanations

closing parenthesis, quotation marks, items

The neuron activates on numeric literals—integers, decimals, and fractional numbers—in the text.

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

 besonderes

-0.87

 holistic

-0.84

 dbName

-0.84

体が

-0.82

:)

-0.82

 companies

-0.80

 dateFormat

-0.79

 several

-0.78

 even

-0.78

roberto

-0.78

POSITIVE LOGITS

and

1.27

unking

1.02

 consommer

0.96

 mélanger

0.93

 procéder

0.91

 جديدة

0.90

 rechercher

0.90

 البحث

0.86

animity

0.86

lisi

0.85

Activations Density 0.161%