INDEX

Explanations

clear or cleared

The neuron fires on occurrences of the substring “clear” (in words like clear, clearing, clearfix, ClearVars, etc.).

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

</h2>

-1.76

 Betrachtung

-1.73

皷

-1.70

 все

-1.69

 Antrags

-1.65

硖

-1.63

 také

-1.61

 Merkmale

-1.61

 Voraussetzung

-1.60

 अब

-1.58

POSITIVE LOGITS

»;

1.78

 baumwolle

1.66

Մ

1.66

лением

1.64

 Clearing

1.58

лению

1.57

蚴

1.56

the

1.55

臯

1.53

 clear

1.52

Activations Density 0.019%