INDEX

Explanations

diverse groups and backgrounds

The neuron activates on mentions of a “diverse” collection—e.g. phrases like “diverse group,” “diverse set,” or “diverse array.”

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

 sollten

-1.70

two

-1.62

髣

-1.52

龶

-1.52

 three

-1.52

or

-1.51

”—

-1.50

满是

-1.49

");

-1.49

 some

-1.49

POSITIVE LOGITS

 profond

1.65

Ⓢ

1.55

 разные

1.44

 позволяет

1.36

 nativo

1.35

diverse

1.34

 возможность

1.34

 nhưng

1.31

carnated

1.31

 posky

1.28

Activations Density 0.037%