INDEX

Explanations

resembling or similarity to

The neuron fires on numeric measurements and quantitative values (especially decimal numbers and percentages).

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

 también

-1.57

 também

-1.43

 også

-1.41

 neuen

-1.39

for

-1.39

 także

-1.38

új

-1.31

 misalnya

-1.30

 همچنین

-1.27

например

-1.23

POSITIVE LOGITS

 other

1.95

 when

1.46

so

1.40

its

1.37

 nativa

1.34

 many

1.30

 still

1.26

 some

1.24

 only

1.19

asks

1.16

Activations Density 0.191%