INDEX

Explanations

plenty of

The neuron detects phrases expressing a large, vague quantity (e.g. “lots of,” “plenty of,” “gobs of”).

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

 olympiques

-1.50

 również

-1.50

去年

-1.46

vų

-1.37

 essas

-1.37

ASTIC

-1.37

 ironically

-1.36

 each

-1.34

 additionally

-1.34

 همچنین

-1.27

POSITIVE LOGITS

1.69

vertebrates

1.50

cipient

1.45

各类

1.45

various

1.44

纔

1.42

end

1.38

{})

1.35

就行了

1.34

嗎

1.33

Activations Density 0.017%