INDEX

Explanations

boring / dull

The neuron fires on words that label something as dull or uninteresting (e.g. “boring,” “blah,” “stale,” 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

 laajenta

-1.34

➬

-1.31

 Focusing

-1.30

 burbu

-1.28

("")){

-1.27

").

-1.24

 représenter

-1.23

ándor

-1.22

 tästä

-1.21

 kaikk

-1.20

POSITIVE LOGITS

 boring

1.73

1.66

or

1.41

骘

1.36

 predictable

1.35

1.34

 dull

1.33

 selectively

1.33

 riche

1.30

 Boring

1.30

Activations Density 0.037%