INDEX

Explanations

screened, aired, or showed

The neuron activates on mentions of a film or TV release/screening/broadcast (e.g. words like “screened,” “airing,” “released,” “shown”).

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

medriver

-1.41

有多少

-1.39

 nawet

-1.31

 که

-1.30

CppCodeGen

-1.28

蕻

-1.25

 גם

-1.24

いたします

-1.23

☫

-1.23

POSITIVE LOGITS

 some

1.64

 valami

1.45

 néhány

1.45

 something

1.38

in

1.36

sprzedam

1.31

 niekoľ

1.30

 Deze

1.28

 okazji

1.24

Larry

1.23

Activations Density 0.053%